What is the upper limit of number of SNP based mutation in any protein?

What is the upper limit of number of SNP based mutation in any protein?

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

I want to know if there is any upper limit of, how many point mutation a protein can have because of disease or nsSNPs? The general studies only focus mostly on a single or double site mutation of any protein structure.

Ka/Ks ratio

In genetics, the Ka/Ks ratio, also known as ω or dN/dS ratio, [a] is used to estimate the balance between neutral mutations, purifying selection and beneficial mutations acting on a set of homologous protein-coding genes. It is calculated as the ratio of the number of nonsynonymous substitutions per non-synonymous site (Ka), in a given period of time, to the number of synonymous substitutions per synonymous site (Ks), in the same period. The latter are assumed to be neutral, so that the ratio indicates the net balance between deleterious and beneficial mutations. Values of Ka/Ks significantly above 1 are unlikely to occur without at least some of the mutations being advantageous. If beneficial mutations are assumed to make little contribution, then Ks estimates the degree of evolutionary constraint.


Immunosuppressive enzymes, which control intracellular and extracellular amino-acid content while simultaneously producing toxic metabolites, have been shown to participate in immune regulation by affecting the proliferative and differentiation capacities of T cells. Their contribution to cancer immune escape has also been revealed, leading to the development of several specific inhibitors of the first identified enzyme of this family, indoleamine 2,3 dioxygenase, now tested in clinical trials.

The most recently described member of this class of enzymes, the phenylalanine oxidase interleukin 4-induced gene 1 (IL4I1) is a secreted protein, which is strongly expressed by infiltrating macrophages in human cancers and in some cases by the malignant cells themselves. 1 Its role in the inhibition of the CD8 + T-cell anti-tumour response has been demonstrated in a melanoma mouse model. 2 IL4I1 presents a strong homology to lower vertebrate l -amino acid oxidases (LAAO), which generally have a lower Km for the amino-acid substrate and a higher Vmax. 3 These proteins use FAD as a cofactor and function as tetramers. 4

Enzymatic activity can be strongly disturbed by protein sequence modifications, which can affect the 3D conformation of the catalytic site, multimer interactions, as well as substrate access to and/or product exit from the catalytic site. For this reason, single-nucleotide polymorphisms (SNPs) or mutations in a gene coding an immunosuppressive enzyme can affect the amino-acid catabolic activity of the protein, thus amplifying or reducing its immune modulatory functions.

In this work, we have compiled the reported SNP and mutation data described for the IL4I1 gene. We have selected and cloned a missense SNP and a mutation, and have verified the functional consequences on IL4I1 enzymatic activity. Three-dimensional modelling of their influence on the structure of the enzyme was performed in an attempt to understand the structural basis for their effect on IL4I1 enzymatic activity.

Mutations in ctDNA

Approaches for the mutation analysis of ctDNA

Mutations in ctDNA from liquid biopsy samples can be detected via two different approaches. In the first approach, single, or low numbers of, mutations can be detected using highly sensitive techniques with high specificity and at a rather fast and cost-effective rate. 22 In 2016, the Cobas EGFR mutation Test v2 that interrogates by RT-PCR several mutations in exons 18, 19, 20 and 21 of epidermal growth factor receptor (EGFR) gene was the first liquid biopsy-based companion diagnostic to be approved by US Food and Drug Administration (FDA) and the European Medicines Agency for the prescription of EGFR inhibitors in patients with non-small-cell lung cancer (NSCLC) in cases when tumour biopsy tissue is not available. 23 Other targeted approaches, based mainly on digital PCR (droplet digital [ddPCR] or BEAMing dPCR), have been demonstrated to be able to detect specific known mutations, such as the main driver mutations of the primary tumour or variants associated with response to drugs in individual tumour types, and usually show high concordance with results obtained in tumour tissue 24,25,26 and reach a variant or mutant allele frequency detection (VAF/MAF) as low as 0.001% for the most advanced technologies 27 (i.e. the frequency of a particular genetic variation of a specific sequence [e.g. allele/mutation] relative to the other genetic variations of the same sequence). The detection and comprehensive molecular characterisation of minimal residual disease (MRD) is of particular importance in the adjuvant setting to improve clinical outcomes 28 ctDNA detected via such targeted, highly sensitive approaches in the early stages of melanoma was reported to predict the relapse risk, 29,30 and might therefore be useful in the process of patient stratification for adjuvant therapy. Next step in the implementation of ctDNA in clinical routine is to demonstrate its utility in patient treatment selection. For instance, in the recently published TARGET study (registered in NIHR Central Portfolio Management System under the reference CPMS ID 39172), the primary aim was to match advanced stage patients to early phase clinical trials on the basis of plasma ctDNA analysis of both somatic mutations and copy number alterations in 641 cancer-associated-genes. 31 Another example is the Circulating Tumour DNA Guided Switch (CAcTUS) study (NCT03808441), which determines whether switching from targeted therapy to immunotherapy based on a decrease in levels of ctDNA in the blood will improve the outcome in melanoma patients.

Broader approaches have also been developed to interrogate multiple mutations in parallel and range from the analysis of several tens of mutations, to a genome-wide analysis of cfDNA by whole-exome sequencing (WES) or whole-genome sequencing (WGS). Most of these approaches use next-generation sequencing (NGS) but mass-spectrometry-based detection of PCR amplicons is also becoming available. 32 Besides increasing the probability of detecting a mutation in cfDNA, these broader approaches allow a more complete genotyping of the tumour, which can be used to assess tumour heterogeneity or to follow clonal evolution of the tumour under treatment, as well as to identify potential resistance mutations before clinical progression is observed. 10,33,34 Another example of the application of nontargeted approaches also relates to cancer patients treated by immunotherapy, for whom mutation load (i.e. the number of nonsynonymous mutations found in a tumour) has emerged as a putative biomarker of the response to the treatment. Assessing mutation load and measuring its evolution through plasma analysis has also been evaluated as an alternative approach to tumour tissue determination. 35,36 More generally, comprehensive reviews have discussed the clinical utility of ctDNA in the new era of immunotherapy. 37,38

However, one should be aware that the larger the panels, the more expensive and difficult it is to obtain high sensitivity for mutation calling.

Challenges associated with mutation detection in cfDNA

A key issue in the analysis of ctDNA is still the extent to which the information gained from the liquid biopsy sample reflects the tumour tissue. Both technical and biological factors can affect the concordance between tumour and plasma, generating false-negative and false-positive results in ctDNA analysis.

False-negative results might be explained by the low volume of plasma yielded (4–5 ml) from a typical blood sample of 10 ml, which limits the total number of available genome copies to be analysed: mutations within a tumour can be clonal or subclonal, and the amount of available genome copies is a limiting factor for the detection of variants of low allele frequency. 39 Moreover, the tumour fraction of cfDNA varies between cancer types as well as between patients affected by the same cancer type. 40 Even at the metastatic stage, some patients can yield a low amount of ctDNA, 41,42 and the question of why some tumours undergo limited shedding of ctDNA is still not completely resolved. In this regard, detection of mitochondrial tumour-derived DNA, as an alternative source of ctDNA might be a promising approach, owing to the thousands of copies of mitochondrial DNA per cell. 43 Proof of principle for this apporach was provided in patient-derived orthotopic xenograft models of glioblastoma in 2019. 11 Considerations about technical improvements for the methods used to analyse cfDNA could also help to overcome the limit of detection. Ultra-deep sequencing methods can lower the percentage of false negative and are currently under evaluation across different cancer types. 44,45,46,47 The size selection of cfDNA fragments (see below) or the choice of an alternative method for library preparation like single strand DNA libraries for NGS are additional solutions. 48

False-positive results are another concerning issue when multiple mutations are interrogated by NGS platforms. The risk of introducing errors during library preparation and subsequent sequencing steps has led to the implementation of multiple mutation-enrichment methods and error-suppression strategies such as the introduction of molecular barcodes or bioinformatic analysis pipelines of the data. 22,39,49 The extensive comparison of paired tumour and plasma samples therefore represents an important prerequisite to evaluate the diagnostic accuracy of analytical platforms, especially for variants with allele fractions that are close to the limit of detection. 50,51,52 Different commercial NGS platforms might not have the same limit of detection or interrogate the same genomic regions as each other, and the field would benefit from rigorous cross-assay comparisons, as carried out between 2015 and 2019 by the EU Innovative Medicines Initiative (IMI) consortium CANCER-ID ( and sustained by the new European Liquid Biopsy Society (ELBS and other networks (the US Blood Profiling Atlas of Cancer A cross-comparison of four commercial NGS platforms, all certified by the US-based college of American Pathologists-Clinical Laboratory Improvement Amendments, was carried out in 2019 with plasma–tumour-matched samples of early stage cancers that present a limited ctDNA amount. 53 Substantial variability in terms of sensitivity (38–89%) and positive predictive values (36–80%) was identified among the different platforms. Low predictive positive values were mainly associated with variants with an allele frequency below 1% and could be explained by technical factors (limited sensitivity, bioinformatic filtering of the data or even plain error of identification). Nonetheless, germline variants shed from normal cells and during clonal haematopoiesis (e.g. the presence of somatic variation in some cancer-related genes like TP53 that do not necessarily lead to cancer) constitute another source of confounding factors that have to be considered when interpreting the data. By applying a highly sensitive and specific ctDNA sequencing assay on a cohort of 124 metastatic cancer patients and 47 controls without cancer, with matched white blood cell DNA, Razavi et al. found that 53.2% of mutations found in cancer patients had features consistent with clonal haematopoiesis. 47 This study highlights therefore the risk of false findings and the need to integrate white blood cell DNA as control when applying ultrasensitive ctDNA sequencing methods. Overall, it appears necessary that laboratories should comment on these different limitations in their reports. 54

If these technical and biological factors could be ruled out, then ctDNA could be used to evaluate intratumour heterogeneity, as it is now well accepted that a single tumour biopsy procedure generates a limited representation of temporal and spatial heterogeneity, whereas ctDNA in plasma would represent a pool of the entire tumour or of the metastatic sites. 55 Up until now, clinical studies that have compared plasma analysis with multiregional tissue biopsies are rare and limited to few patients, due to an increase risk of clinical adverse side effects linked to this invasive procedure (see Table 1). In this sense, studies conducted utilising rapid autopsy programs are of particular interest. 26 Some studies have shown that the quantitative level of mutations found in ctDNA reflects the architecture of the mutational landscape in tumour tissue, with truncal mutations more readily detectable than private mutations. 10,56,57,58 In the context of acquired resistance in gastrointestinal cancers, mutation analysis of ctDNA taken at progression was more informative than the corresponding analysis of tissue biopsies. 34 However, in some cases of melanoma patients ctDNA analysis only partially reflected heterogeneity, with under-representation of certain anatomical metastatic sites like brain or subcutaneous metastases. 12 A better understanding of the parameters that govern ctDNA release (i.e. proliferation/turnover, active secretion, type of cancer, location or tumour vascularity) is therefore needed.


For many mutants, using traditional, map-based positional cloning is an extremely difficult approach for the identification of the genetic basis of some phenotypes. Here, we demonstrated the utility of massively parallel sequencing using an ABI SOLiD sequencer to spot EMS-induced mutations in a non-reference strain of Arabidopsis. Using a functional genomic approach, based on the assumption that a clock component gene is likely to be rhythmically expressed, we were able to further narrow down the number of candidate SNPs. Finally, by using the SNP information we were able to exclude the previously identified clock gene PRR7 by generating clean backcrossed lines, identifying a SNP in the gene AtNFXL-2 as the likely cause of the ebi-1 phenotype. This was further validated by the characterization of a second allele of ebi, ebi-2. Our approach demonstrates the feasibility of next generation sequencing as a tool for positionally cloning genes in a large genome.

The gene responsible for the ebi-1 phenotype, AtNFXL-2, is a zinc finger transcription factor, a homolog of the human NF-X1 protein. In humans, NF-X1 binds to the X-box found in class II MHC genes [29]. Arabidopsis has two NF-X1 homologs, AtNFXL-1 and AtNFXL-2, which are thought to act antagonistically to regulate genes involved in salt, osmotic and drought stress, with AtNFXL-1 activating and AtNFXL-2 repressing stress-inducing genes [30]. AtNFXL-1 has also been suggested to be a negative regulator of defense-related genes [31] and temperature stress [32]. Thus, the clock phenotype of the AtNFXL-2 mutant provides an intriguing link between the clock and biotic and abiotic stress responses. This link has already been alluded to in a recent review [33] and in the identification of a possible role for the clock protein GI in cold stress tolerance [34].

Critical to the success of this project was to sequence the original parent from which the EMS mutant was derived. When Col-0 was recently re-sequenced using a lab strain, 1,172 SNPs were identified between the lab strain Col-0 and the original reference genome of Col-0. It is clear, therefore, that sequencing the original parent rather than relying on a previously sequenced reference is the correct approach. Secondly, the fact that we used a backcrossed line reduced the number of EMS mutations we had to consider from approximately 1,200 to 109. The large number of 'piggy-backing' SNPs also provides a stark example of just how many non-synonymous/nonsense mutations (51) are still present in what is regarded by the community as a 'clean' line.

An alternative approach to the direct sequencing method described here has been reported [16, 17]. The technique relies on accurately scoring mutant individuals in an F2 mapping cross between divergent Arabidopsis accessions and then combining these individuals and sequencing the bulked DNA using next generation sequencing. The output of the sequence data provides information about the mapping position and a number of candidate SNPs. While this approach is extremely valuable, where the phenotype is subtle and there is a large amount of phenotype variation between individuals (resulting in a high number of false positives) it is unlikely to be useful. For the ebi-1 mutant, mapping was only possible by re-scoring potential mutants isolated in F2 again in the F3.

Our data clearly indicate strand bias in the mutagenesis process, resulting in long series of C to T or G to A transitions, rather than random mutation of either strand as expected based on previous population-level investigations [22]. It has been shown that transcriptional activity affects repair efficiency [35],, although this is unlikely to explain the bias, as over the long stretches of genome, both strands of the DNA are transcriptionally active. One simple explanation is that the mutagenesis event occurs and each strand of DNA is replicated and segregates to separate daughter cells. This would be sufficient to confer strand bias and thus the long stretches of identical transitions.

This combined approach of next generation sequencing and functional genomics can be used to identify genes previously intractable to conventional mapping approaches. The methodology is not restricted to Arabidopsis or to EMS-induced SNPs, but could be used to positionally clone genes in any organism with a sequenced genome. As accuracy and throughput increases, the technique should be possible in larger more complex genomes.


The first human genome sequences were published in nearly complete draft form in February 2001 by the Human Genome Project [15] and Celera Corporation. [16] Completion of the Human Genome Project's sequencing effort was announced in 2004 with the publication of a draft genome sequence, leaving just 341 gaps in the sequence, representing highly-repetitive and other DNA that could not be sequenced with the technology available at the time. [8] The human genome was the first of all vertebrates to be sequenced to such near-completion, and as of 2018, the diploid genomes of over a million individual humans had been determined using next-generation sequencing. [17] In 2021 it was reported that the T2T consortium had filled in all of the gaps. Thus there came into existence a complete human genome with no gaps. [18]

These data are used worldwide in biomedical science, anthropology, forensics and other branches of science. Such genomic studies have led to advances in the diagnosis and treatment of diseases, and to new insights in many fields of biology, including human evolution.

In June 2016, scientists formally announced HGP-Write, a plan to synthesize the human genome. [19] [20]

Although the 'completion' of the human genome project was announced in 2001, [14] there remained hundreds of gaps, with about 5–10% of the total sequence remaining undetermined. The missing genetic information was mostly in repetitive heterochromatic regions and near the centromeres and telomeres, but also some gene-encoding euchromatic regions. [21] There remained 160 euchromatic gaps in 2015 when the sequences spanning another 50 formerly-unsequenced regions were determined. [22] Only in 2020 was the first truly complete telomere-to-telomere sequence of a human chromosome determined, namely of the X chromosome. [23]

The total length of the human reference genome, that does not represent the sequence of any specific individual, is over 3 billion base pairs. The genome is organized into 22 paired chromosomes, termed autosomes, plus the 23rd pair of sex chromosomes (XX) in the female, and (XY) in the male. These are all large linear DNA molecules contained within the cell nucleus. The genome also includes the mitochondrial DNA, a comparatively small circular molecule present in multiple copies in each the mitochondrion.

Human reference genome data, by chromosome [24]
Chromosome Length
Variations Protein-
miRNA rRNA snRNA snoRNA Misc
Links Centromere
1 85 248,956,422 12,151,146 2058 1220 1200 496 134 66 221 145 192 EBI 125 7.9
2 83 242,193,529 12,945,965 1309 1023 1037 375 115 40 161 117 176 EBI 93.3 16.2
3 67 198,295,559 10,638,715 1078 763 711 298 99 29 138 87 134 EBI 91 23
4 65 190,214,555 10,165,685 752 727 657 228 92 24 120 56 104 EBI 50.4 29.6
5 62 181,538,259 9,519,995 876 721 844 235 83 25 106 61 119 EBI 48.4 35.8
6 58 170,805,979 9,130,476 1048 801 639 234 81 26 111 73 105 EBI 61 41.6
7 54 159,345,973 8,613,298 989 885 605 208 90 24 90 76 143 EBI 59.9 47.1
8 50 145,138,636 8,221,520 677 613 735 214 80 28 86 52 82 EBI 45.6 52
9 48 138,394,717 6,590,811 786 661 491 190 69 19 66 51 96 EBI 49 56.3
10 46 133,797,422 7,223,944 733 568 579 204 64 32 87 56 89 EBI 40.2 60.9
11 46 135,086,622 7,535,370 1298 821 710 233 63 24 74 76 97 EBI 53.7 65.4
12 45 133,275,309 7,228,129 1034 617 848 227 72 27 106 62 115 EBI 35.8 70
13 39 114,364,328 5,082,574 327 372 397 104 42 16 45 34 75 EBI 17.9 73.4
14 36 107,043,718 4,865,950 830 523 533 239 92 10 65 97 79 EBI 17.6 76.4
15 35 101,991,189 4,515,076 613 510 639 250 78 13 63 136 93 EBI 19 79.3
16 31 90,338,345 5,101,702 873 465 799 187 52 32 53 58 51 EBI 36.6 82
17 28 83,257,441 4,614,972 1197 531 834 235 61 15 80 71 99 EBI 24 84.8
18 27 80,373,285 4,035,966 270 247 453 109 32 13 51 36 41 EBI 17.2 87.4
19 20 58,617,616 3,858,269 1472 512 628 179 110 13 29 31 61 EBI 26.5 89.3
20 21 64,444,167 3,439,621 544 249 384 131 57 15 46 37 68 EBI 27.5 91.4
21 16 46,709,983 2,049,697 234 185 305 71 16 5 21 19 24 EBI 13.2 92.6
22 17 50,818,468 2,135,311 488 324 357 78 31 5 23 23 62 EBI 14.7 93.8
X 53 156,040,895 5,753,881 842 874 271 258 128 22 85 64 100 EBI 60.6 99.1
Y 20 57,227,415 211,643 71 388 71 30 15 7 17 3 8 EBI 10.4 100
mtDNA 0.0054 16,569 929 13 0 0 24 0 2 0 0 0 EBI N/A 100
total 3,088,286,401 155,630,645 20412 14600 14727 5037 1756 532 1944 1521 2213

Original analysis published in the Ensembl database at the European Bioinformatics Institute (EBI) and Wellcome Trust Sanger Institute. Chromosome lengths estimated by multiplying the number of base pairs by 0.34 nanometers (distance between base pairs in the most common structure of the DNA double helix a recent estimate of human chromosome lengths based on updated data reports 205.00 cm for the diploid male genome and 208.23 cm for female, corresponding to weights of 6.41 and 6.51 picograms (pg), respectively [25] ). Number of proteins is based on the number of initial precursor mRNA transcripts, and does not include products of alternative pre-mRNA splicing, or modifications to protein structure that occur after translation.

Variations are unique DNA sequence differences that have been identified in the individual human genome sequences analyzed by Ensembl as of December 2016. The number of identified variations is expected to increase as further personal genomes are sequenced and analyzed. In addition to the gene content shown in this table, a large number of non-expressed functional sequences have been identified throughout the human genome (see below). Links open windows to the reference chromosome sequences in the EBI genome browser.

Small non-coding RNAs are RNAs of as many as 200 bases that do not have protein-coding potential. These include: microRNAs, or miRNAs (post-transcriptional regulators of gene expression), small nuclear RNAs, or snRNAs (the RNA components of spliceosomes), and small nucleolar RNAs, or snoRNA (involved in guiding chemical modifications to other RNA molecules). Long non-coding RNAs are RNA molecules longer than 200 bases that do not have protein-coding potential. These include: ribosomal RNAs, or rRNAs (the RNA components of ribosomes), and a variety of other long RNAs that are involved in regulation of gene expression, epigenetic modifications of DNA nucleotides and histone proteins, and regulation of the activity of protein-coding genes. Small discrepancies between total-small-ncRNA numbers and the numbers of specific types of small ncNRAs result from the former values being sourced from Ensembl release 87 and the latter from Ensembl release 68.

The number of genes in the human genome is not entirely clear because the function of numerous transcripts remains unclear. This is especially true for non-coding RNA. The number of protein-coding genes is better known but there are still on the order of 1,400 questionable genes which may or may not encode functional proteins, usually encoded by short open reading frames.

Discrepancies in human gene number estimates among different databases, as of July 2018 [26]
Gencode [27] Ensembl [28] Refseq [29] CHESS [30]
protein-coding genes 19,901 20,376 20,345 21,306
lncRNA genes 15,779 14,720 17,712 18,484
antisense RNA 5501 28 2694
miscellaneous RNA 2213 2222 13,899 4347
Pseudogenes 14,723 1740 15,952
total transcripts 203,835 203,903 154,484 328,827

Information content Edit

The haploid human genome (23 chromosomes) is about 3 billion base pairs long and contains around 30,000 genes. [31] Since every base pair can be coded by 2 bits, this is about 750 megabytes of data. An individual somatic (diploid) cell contains twice this amount, that is, about 6 billion base pairs. Men have fewer than women because the Y chromosome is about 57 million base pairs whereas the X is about 156 million. Since individual genomes vary in sequence by less than 1% from each other, the variations of a given human's genome from a common reference can be losslessly compressed to roughly 4 megabytes. [32]

The entropy rate of the genome differs significantly between coding and non-coding sequences. It is close to the maximum of 2 bits per base pair for the coding sequences (about 45 million base pairs), but less for the non-coding parts. It ranges between 1.5 and 1.9 bits per base pair for the individual chromosome, except for the Y-chromosome, which has an entropy rate below 0.9 bits per base pair. [33]

The content of the human genome is commonly divided into coding and noncoding DNA sequences. Coding DNA is defined as those sequences that can be transcribed into mRNA and translated into proteins during the human life cycle these sequences occupy only a small fraction of the genome (<2%). Noncoding DNA is made up of all of those sequences (ca. 98% of the genome) that are not used to encode proteins.

Some noncoding DNA contains genes for RNA molecules with important biological functions (noncoding RNA, for example ribosomal RNA and transfer RNA). The exploration of the function and evolutionary origin of noncoding DNA is an important goal of contemporary genome research, including the ENCODE (Encyclopedia of DNA Elements) project, which aims to survey the entire human genome, using a variety of experimental tools whose results are indicative of molecular activity.

Because non-coding DNA greatly outnumbers coding DNA, the concept of the sequenced genome has become a more focused analytical concept than the classical concept of the DNA-coding gene. [34] [35]

Protein-coding sequences represent the most widely studied and best understood component of the human genome. These sequences ultimately lead to the production of all human proteins, although several biological processes (e.g. DNA rearrangements and alternative pre-mRNA splicing) can lead to the production of many more unique proteins than the number of protein-coding genes. The complete modular protein-coding capacity of the genome is contained within the exome, and consists of DNA sequences encoded by exons that can be translated into proteins. Because of its biological importance, and the fact that it constitutes less than 2% of the genome, sequencing of the exome was the first major milepost of the Human Genome Project.

Number of protein-coding genes. About 20,000 human proteins have been annotated in databases such as Uniprot. [37] Historically, estimates for the number of protein genes have varied widely, ranging up to 2,000,000 in the late 1960s, [38] but several researchers pointed out in the early 1970s that the estimated mutational load from deleterious mutations placed an upper limit of approximately 40,000 for the total number of functional loci (this includes protein-coding and functional non-coding genes). [39] The number of human protein-coding genes is not significantly larger than that of many less complex organisms, such as the roundworm and the fruit fly. This difference may result from the extensive use of alternative pre-mRNA splicing in humans, which provides the ability to build a very large number of modular proteins through the selective incorporation of exons.

Protein-coding capacity per chromosome. Protein-coding genes are distributed unevenly across the chromosomes, ranging from a few dozen to more than 2000, with an especially high gene density within chromosomes 1, 11, and 19. Each chromosome contains various gene-rich and gene-poor regions, which may be correlated with chromosome bands and GC-content. [40] The significance of these nonrandom patterns of gene density is not well understood. [41]

Size of protein-coding genes. The size of protein-coding genes within the human genome shows enormous variability. For example, the gene for histone H1a (HIST1HIA) is relatively small and simple, lacking introns and encoding an 781 nucleotide-long mRNA that produces a 215 amino acid protein from its 648 nucleotide open reading frame. Dystrophin (DMD) was the largest protein-coding gene in the 2001 human reference genome, spanning a total of 2.2 million nucleotides, [42] while more recent systematic meta-analysis of updated human genome data identified an even larger protein-coding gene, RBFOX1 (RNA binding protein, fox-1 homolog 1), spanning a total of 2.47 million nucleotides. [43] Titin (TTN) has the longest coding sequence (114,414 nucleotides), the largest number of exons (363), [42] and the longest single exon (17,106 nucleotides). As estimated based on a curated set of protein-coding genes over the whole genome, the median size is 26,288 nucleotides (mean = 66,577), the median exon size, 133 nucleotides (mean = 309), the median number of exons, 8 (mean = 11), and the median encoded protein is 425 amino acids (mean = 553) in length. [43]

Examples of human protein-coding genes [44]
Protein Chrom Gene Length Exons Exon length Intron length Alt splicing
Breast cancer type 2 susceptibility protein 13 BRCA2 83,736 27 11,386 72,350 yes
Cystic fibrosis transmembrane conductance regulator 7 CFTR 202,881 27 4,440 198,441 yes
Cytochrome b MT MTCYB 1,140 1 1,140 0 no
Dystrophin X DMD 2,220,381 79 10,500 2,209,881 yes
Glyceraldehyde-3-phosphate dehydrogenase 12 GAPDH 4,444 9 1,425 3,019 yes
Hemoglobin beta subunit 11 HBB 1,605 3 626 979 no
Histone H1A 6 HIST1H1A 781 1 781 0 no
Titin 2 TTN 281,434 364 104,301 177,133 yes

Noncoding DNA is defined as all of the DNA sequences within a genome that are not found within protein-coding exons, and so are never represented within the amino acid sequence of expressed proteins. By this definition, more than 98% of the human genomes is composed of ncDNA.

Numerous classes of noncoding DNA have been identified, including genes for noncoding RNA (e.g. tRNA and rRNA), pseudogenes, introns, untranslated regions of mRNA, regulatory DNA sequences, repetitive DNA sequences, and sequences related to mobile genetic elements.

Numerous sequences that are included within genes are also defined as noncoding DNA. These include genes for noncoding RNA (e.g. tRNA, rRNA), and untranslated components of protein-coding genes (e.g. introns, and 5' and 3' untranslated regions of mRNA).

Protein-coding sequences (specifically, coding exons) constitute less than 1.5% of the human genome. [14] In addition, about 26% of the human genome is introns. [45] Aside from genes (exons and introns) and known regulatory sequences (8–20%), the human genome contains regions of noncoding DNA. The exact amount of noncoding DNA that plays a role in cell physiology has been hotly debated. Recent analysis by the ENCODE project indicates that 80% of the entire human genome is either transcribed, binds to regulatory proteins, or is associated with some other biochemical activity. [12]

It however remains controversial whether all of this biochemical activity contributes to cell physiology, or whether a substantial portion of this is the result transcriptional and biochemical noise, which must be actively filtered out by the organism. [46] Excluding protein-coding sequences, introns, and regulatory regions, much of the non-coding DNA is composed of: Many DNA sequences that do not play a role in gene expression have important biological functions. Comparative genomics studies indicate that about 5% of the genome contains sequences of noncoding DNA that are highly conserved, sometimes on time-scales representing hundreds of millions of years, implying that these noncoding regions are under strong evolutionary pressure and positive selection. [47]

Many of these sequences regulate the structure of chromosomes by limiting the regions of heterochromatin formation and regulating structural features of the chromosomes, such as the telomeres and centromeres. Other noncoding regions serve as origins of DNA replication. Finally several regions are transcribed into functional noncoding RNA that regulate the expression of protein-coding genes (for example [48] ), mRNA translation and stability (see miRNA), chromatin structure (including histone modifications, for example [49] ), DNA methylation (for example [50] ), DNA recombination (for example [51] ), and cross-regulate other noncoding RNAs (for example [52] ). It is also likely that many transcribed noncoding regions do not serve any role and that this transcription is the product of non-specific RNA Polymerase activity. [46]

Pseudogenes Edit

Pseudogenes are inactive copies of protein-coding genes, often generated by gene duplication, that have become nonfunctional through the accumulation of inactivating mutations. The number of pseudogenes in the human genome is on the order of 13,000, [53] and in some chromosomes is nearly the same as the number of functional protein-coding genes. Gene duplication is a major mechanism through which new genetic material is generated during molecular evolution.

For example, the olfactory receptor gene family is one of the best-documented examples of pseudogenes in the human genome. More than 60 percent of the genes in this family are non-functional pseudogenes in humans. By comparison, only 20 percent of genes in the mouse olfactory receptor gene family are pseudogenes. Research suggests that this is a species-specific characteristic, as the most closely related primates all have proportionally fewer pseudogenes. This genetic discovery helps to explain the less acute sense of smell in humans relative to other mammals. [54]

Genes for noncoding RNA (ncRNA) Edit

Noncoding RNA molecules play many essential roles in cells, especially in the many reactions of protein synthesis and RNA processing. Noncoding RNA include tRNA, ribosomal RNA, microRNA, snRNA and other non-coding RNA genes including about 60,000 long non-coding RNAs (lncRNAs). [12] [55] [56] [57] Although the number of reported lncRNA genes continues to rise and the exact number in the human genome is yet to be defined, many of them are argued to be non-functional. [58]

Many ncRNAs are critical elements in gene regulation and expression. Noncoding RNA also contributes to epigenetics, transcription, RNA splicing, and the translational machinery. The role of RNA in genetic regulation and disease offers a new potential level of unexplored genomic complexity. [59]

Introns and untranslated regions of mRNA Edit

In addition to the ncRNA molecules that are encoded by discrete genes, the initial transcripts of protein coding genes usually contain extensive noncoding sequences, in the form of introns, 5'-untranslated regions (5'-UTR), and 3'-untranslated regions (3'-UTR). Within most protein-coding genes of the human genome, the length of intron sequences is 10- to 100-times the length of exon sequences.

Regulatory DNA sequences Edit

The human genome has many different regulatory sequences which are crucial to controlling gene expression. Conservative estimates indicate that these sequences make up 8% of the genome, [60] however extrapolations from the ENCODE project give that 20 [61] -40% [62] of the genome is gene regulatory sequence. Some types of non-coding DNA are genetic "switches" that do not encode proteins, but do regulate when and where genes are expressed (called enhancers). [63]

Regulatory sequences have been known since the late 1960s. [64] The first identification of regulatory sequences in the human genome relied on recombinant DNA technology. [65] Later with the advent of genomic sequencing, the identification of these sequences could be inferred by evolutionary conservation. The evolutionary branch between the primates and mouse, for example, occurred 70–90 million years ago. [66] So computer comparisons of gene sequences that identify conserved non-coding sequences will be an indication of their importance in duties such as gene regulation. [67]

Other genomes have been sequenced with the same intention of aiding conservation-guided methods, for exampled the pufferfish genome. [68] However, regulatory sequences disappear and re-evolve during evolution at a high rate. [69] [70] [71]

As of 2012, the efforts have shifted toward finding interactions between DNA and regulatory proteins by the technique ChIP-Seq, or gaps where the DNA is not packaged by histones (DNase hypersensitive sites), both of which tell where there are active regulatory sequences in the investigated cell type. [60]

Repetitive DNA sequences Edit

Repetitive DNA sequences comprise approximately 50% of the human genome. [72]

About 8% of the human genome consists of tandem DNA arrays or tandem repeats, low complexity repeat sequences that have multiple adjacent copies (e.g. "CAGCAGCAG. "). [73] The tandem sequences may be of variable lengths, from two nucleotides to tens of nucleotides. These sequences are highly variable, even among closely related individuals, and so are used for genealogical DNA testing and forensic DNA analysis. [74]

Repeated sequences of fewer than ten nucleotides (e.g. the dinucleotide repeat (AC)n) are termed microsatellite sequences. Among the microsatellite sequences, trinucleotide repeats are of particular importance, as sometimes occur within coding regions of genes for proteins and may lead to genetic disorders. For example, Huntington's disease results from an expansion of the trinucleotide repeat (CAG)n within the Huntingtin gene on human chromosome 4. Telomeres (the ends of linear chromosomes) end with a microsatellite hexanucleotide repeat of the sequence (TTAGGG)n.

Tandem repeats of longer sequences (arrays of repeated sequences 10–60 nucleotides long) are termed minisatellites.

Mobile genetic elements (transposons) and their relics Edit

Transposable genetic elements, DNA sequences that can replicate and insert copies of themselves at other locations within a host genome, are an abundant component in the human genome. The most abundant transposon lineage, Alu, has about 50,000 active copies, [75] and can be inserted into intragenic and intergenic regions. [76] One other lineage, LINE-1, has about 100 active copies per genome (the number varies between people). [77] Together with non-functional relics of old transposons, they account for over half of total human DNA. [78] Sometimes called "jumping genes", transposons have played a major role in sculpting the human genome. Some of these sequences represent endogenous retroviruses, DNA copies of viral sequences that have become permanently integrated into the genome and are now passed on to succeeding generations.

Mobile elements within the human genome can be classified into LTR retrotransposons (8.3% of total genome), SINEs (13.1% of total genome) including Alu elements, LINEs (20.4% of total genome), SVAs and Class II DNA transposons (2.9% of total genome).

Human reference genome Edit

With the exception of identical twins, all humans show significant variation in genomic DNA sequences. The human reference genome (HRG) is used as a standard sequence reference.

There are several important points concerning the human reference genome:

  • The HRG is a haploid sequence. Each chromosome is represented once.
  • The HRG is a composite sequence, and does not correspond to any actual human individual.
  • The HRG is periodically updated to correct errors, ambiguities, and unknown "gaps".
  • The HRG in no way represents an "ideal" or "perfect" human individual. It is simply a standardized representation or model that is used for comparative purposes.

The Genome Reference Consortium is responsible for updating the HRG. Version 38 was released in December 2013. [79]

Measuring human genetic variation Edit

Most studies of human genetic variation have focused on single-nucleotide polymorphisms (SNPs), which are substitutions in individual bases along a chromosome. Most analyses estimate that SNPs occur 1 in 1000 base pairs, on average, in the euchromatic human genome, although they do not occur at a uniform density. Thus follows the popular statement that "we are all, regardless of race, genetically 99.9% the same", [80] although this would be somewhat qualified by most geneticists. For example, a much larger fraction of the genome is now thought to be involved in copy number variation. [81] A large-scale collaborative effort to catalog SNP variations in the human genome is being undertaken by the International HapMap Project.

The genomic loci and length of certain types of small repetitive sequences are highly variable from person to person, which is the basis of DNA fingerprinting and DNA paternity testing technologies. The heterochromatic portions of the human genome, which total several hundred million base pairs, are also thought to be quite variable within the human population (they are so repetitive and so long that they cannot be accurately sequenced with current technology). These regions contain few genes, and it is unclear whether any significant phenotypic effect results from typical variation in repeats or heterochromatin.

Most gross genomic mutations in gamete germ cells probably result in inviable embryos however, a number of human diseases are related to large-scale genomic abnormalities. Down syndrome, Turner Syndrome, and a number of other diseases result from nondisjunction of entire chromosomes. Cancer cells frequently have aneuploidy of chromosomes and chromosome arms, although a cause and effect relationship between aneuploidy and cancer has not been established.

Mapping human genomic variation Edit

Whereas a genome sequence lists the order of every DNA base in a genome, a genome map identifies the landmarks. A genome map is less detailed than a genome sequence and aids in navigating around the genome. [82] [83]

An example of a variation map is the HapMap being developed by the International HapMap Project. The HapMap is a haplotype map of the human genome, "which will describe the common patterns of human DNA sequence variation." [84] It catalogs the patterns of small-scale variations in the genome that involve single DNA letters, or bases.

Researchers published the first sequence-based map of large-scale structural variation across the human genome in the journal Nature in May 2008. [85] [86] Large-scale structural variations are differences in the genome among people that range from a few thousand to a few million DNA bases some are gains or losses of stretches of genome sequence and others appear as re-arrangements of stretches of sequence. These variations include differences in the number of copies individuals have of a particular gene, deletions, translocations and inversions.

Structural variation Edit

Structural variation refers to genetic variants that affect larger segments of the human genome, as opposed to point mutations. Often, structural variants (SVs) are defined as variants of 50 base pairs (bp) or greater, such as deletions, duplications, insertions, inversions and other rearrangements. About 90% of structural variants are noncoding deletions but most individuals have more than a thousand such deletions the size of deletions ranges from dozens of base pairs to tens of thousands of bp. [87] On average, individuals carry

3 rare structural variants that alter coding regions, e.g. delete exons. About 2% of individuals carry ultra-rare megabase-scale structural variants, especially rearrangements. That is, millions of base pairs may be inverted within a chromosome ultra-rare means that they are only found in individuals or their family members and thus have arisen very recently. [87]

SNP frequency across the human genome Edit

Single-nucleotide polymorphisms (SNPs) do not occur homogeneously across the human genome. In fact, there is enormous diversity in SNP frequency between genes, reflecting different selective pressures on each gene as well as different mutation and recombination rates across the genome. However, studies on SNPs are biased towards coding regions, the data generated from them are unlikely to reflect the overall distribution of SNPs throughout the genome. Therefore, the SNP Consortium protocol was designed to identify SNPs with no bias towards coding regions and the Consortium's 100,000 SNPs generally reflect sequence diversity across the human chromosomes. The SNP Consortium aims to expand the number of SNPs identified across the genome to 300 000 by the end of the first quarter of 2001. [88]

Changes in non-coding sequence and synonymous changes in coding sequence are generally more common than non-synonymous changes, reflecting greater selective pressure reducing diversity at positions dictating amino acid identity. Transitional changes are more common than transversions, with CpG dinucleotides showing the highest mutation rate, presumably due to deamination.

Personal genomes Edit

A personal genome sequence is a (nearly) complete sequence of the chemical base pairs that make up the DNA of a single person. Because medical treatments have different effects on different people due to genetic variations such as single-nucleotide polymorphisms (SNPs), the analysis of personal genomes may lead to personalized medical treatment based on individual genotypes. [89]

The first personal genome sequence to be determined was that of Craig Venter in 2007. Personal genomes had not been sequenced in the public Human Genome Project to protect the identity of volunteers who provided DNA samples. That sequence was derived from the DNA of several volunteers from a diverse population. [90] However, early in the Venter-led Celera Genomics genome sequencing effort the decision was made to switch from sequencing a composite sample to using DNA from a single individual, later revealed to have been Venter himself. Thus the Celera human genome sequence released in 2000 was largely that of one man. Subsequent replacement of the early composite-derived data and determination of the diploid sequence, representing both sets of chromosomes, rather than a haploid sequence originally reported, allowed the release of the first personal genome. [91] In April 2008, that of James Watson was also completed. In 2009, Stephen Quake published his own genome sequence derived from a sequencer of his own design, the Heliscope. [92] A Stanford team led by Euan Ashley published a framework for the medical interpretation of human genomes implemented on Quake’s genome and made whole genome-informed medical decisions for the first time. [93] That team further extended the approach to the West family, the first family sequenced as part of Illumina’s Personal Genome Sequencing program. [94] Since then hundreds of personal genome sequences have been released, [95] including those of Desmond Tutu, [96] [97] and of a Paleo-Eskimo. [98] In 2012, the whole genome sequences of two family trios among 1092 genomes was made public. [3] In November 2013, a Spanish family made four personal exome datasets (about 1% of the genome) publicly available under a Creative Commons public domain license. [99] [100] The Personal Genome Project (started in 2005) is among the few to make both genome sequences and corresponding medical phenotypes publicly available. [101] [102]

The sequencing of individual genomes further unveiled levels of genetic complexity that had not been appreciated before. Personal genomics helped reveal the significant level of diversity in the human genome attributed not only to SNPs but structural variations as well. However, the application of such knowledge to the treatment of disease and in the medical field is only in its very beginnings. [103] Exome sequencing has become increasingly popular as a tool to aid in diagnosis of genetic disease because the exome contributes only 1% of the genomic sequence but accounts for roughly 85% of mutations that contribute significantly to disease. [104]

Human knockouts Edit

In humans, gene knockouts naturally occur as heterozygous or homozygous loss-of-function gene knockouts. These knockouts are often difficult to distinguish, especially within heterogeneous genetic backgrounds. They are also difficult to find as they occur in low frequencies.

Populations with high rates of consanguinity, such as countries with high rates of first-cousin marriages, display the highest frequencies of homozygous gene knockouts. Such populations include Pakistan, Iceland, and Amish populations. These populations with a high level of parental-relatedness have been subjects of human knock out research which has helped to determine the function of specific genes in humans. By distinguishing specific knockouts, researchers are able to use phenotypic analyses of these individuals to help characterize the gene that has been knocked out.

Knockouts in specific genes can cause genetic diseases, potentially have beneficial effects, or even result in no phenotypic effect at all. However, determining a knockout's phenotypic effect and in humans can be challenging. Challenges to characterizing and clinically interpreting knockouts include difficulty calling of DNA variants, determining disruption of protein function (annotation), and considering the amount of influence mosaicism has on the phenotype. [105]

One major study that investigated human knockouts is the Pakistan Risk of Myocardial Infarction study. It was found that individuals possessing a heterozygous loss-of-function gene knockout for the APOC3 gene had lower triglycerides in the blood after consuming a high fat meal as compared to individuals without the mutation. However, individuals possessing homozygous loss-of-function gene knockouts of the APOC3 gene displayed the lowest level of triglycerides in the blood after the fat load test, as they produce no functional APOC3 protein. [106]

Most aspects of human biology involve both genetic (inherited) and non-genetic (environmental) factors. Some inherited variation influences aspects of our biology that are not medical in nature (height, eye color, ability to taste or smell certain compounds, etc.). Moreover, some genetic disorders only cause disease in combination with the appropriate environmental factors (such as diet). With these caveats, genetic disorders may be described as clinically defined diseases caused by genomic DNA sequence variation. In the most straightforward cases, the disorder can be associated with variation in a single gene. For example, cystic fibrosis is caused by mutations in the CFTR gene and is the most common recessive disorder in caucasian populations with over 1,300 different mutations known. [107]

Disease-causing mutations in specific genes are usually severe in terms of gene function and are fortunately rare, thus genetic disorders are similarly individually rare. However, since there are many genes that can vary to cause genetic disorders, in aggregate they constitute a significant component of known medical conditions, especially in pediatric medicine. Molecularly characterized genetic disorders are those for which the underlying causal gene has been identified. Currently there are approximately 2,200 such disorders annotated in the OMIM database. [107]

Studies of genetic disorders are often performed by means of family-based studies. In some instances, population based approaches are employed, particularly in the case of so-called founder populations such as those in Finland, French-Canada, Utah, Sardinia, etc. Diagnosis and treatment of genetic disorders are usually performed by a geneticist-physician trained in clinical/medical genetics. The results of the Human Genome Project are likely to provide increased availability of genetic testing for gene-related disorders, and eventually improved treatment. Parents can be screened for hereditary conditions and counselled on the consequences, the probability of inheritance, and how to avoid or ameliorate it in their offspring.

There are many different kinds of DNA sequence variation, ranging from complete extra or missing chromosomes down to single nucleotide changes. It is generally presumed that much naturally occurring genetic variation in human populations is phenotypically neutral, i.e., has little or no detectable effect on the physiology of the individual (although there may be fractional differences in fitness defined over evolutionary time frames). Genetic disorders can be caused by any or all known types of sequence variation. To molecularly characterize a new genetic disorder, it is necessary to establish a causal link between a particular genomic sequence variant and the clinical disease under investigation. Such studies constitute the realm of human molecular genetics.

With the advent of the Human Genome and International HapMap Project, it has become feasible to explore subtle genetic influences on many common disease conditions such as diabetes, asthma, migraine, schizophrenia, etc. Although some causal links have been made between genomic sequence variants in particular genes and some of these diseases, often with much publicity in the general media, these are usually not considered to be genetic disorders per se as their causes are complex, involving many different genetic and environmental factors. Thus there may be disagreement in particular cases whether a specific medical condition should be termed a genetic disorder.

Additional genetic disorders of mention are Kallman syndrome and Pfeiffer syndrome (gene FGFR1), Fuchs corneal dystrophy (gene TCF4), Hirschsprung's disease (genes RET and FECH), Bardet-Biedl syndrome 1 (genes CCDC28B and BBS1), Bardet-Biedl syndrome 10 (gene BBS10), and facioscapulohumeral muscular dystrophy type 2 (genes D4Z4 and SMCHD1). [108]

Genome sequencing is now able to narrow the genome down to specific locations to more accurately find mutations that will result in a genetic disorder. Copy number variants (CNVs) and single nucleotide variants (SNVs) are also able to be detected at the same time as genome sequencing with newer sequencing procedures available, called Next Generation Sequencing (NGS). This only analyzes a small portion of the genome, around 1-2%. The results of this sequencing can be used for clinical diagnosis of a genetic condition, including Usher syndrome, retinal disease, hearing impairments, diabetes, epilepsy, Leigh disease, hereditary cancers, neuromuscular diseases, primary immunodeficiencies, severe combined immunodeficiency (SCID), and diseases of the mitochondria. [109] NGS can also be used to identify carriers of diseases before conception. The diseases that can be detected in this sequencing include Tay-Sachs disease, Bloom syndrome, Gaucher disease, Canavan disease, familial dysautonomia, cystic fibrosis, spinal muscular atrophy, and fragile-X syndrome. The Next Genome Sequencing can be narrowed down to specifically look for diseases more prevalent in certain ethnic populations. [110]

1:15000 in American Caucasians

1:176 in Mennonite/Amish communities

Comparative genomics studies of mammalian genomes suggest that approximately 5% of the human genome has been conserved by evolution since the divergence of extant lineages approximately 200 million years ago, containing the vast majority of genes. [111] [112] The published chimpanzee genome differs from that of the human genome by 1.23% in direct sequence comparisons. [113] Around 20% of this figure is accounted for by variation within each species, leaving only

1.06% consistent sequence divergence between humans and chimps at shared genes. [114] This nucleotide by nucleotide difference is dwarfed, however, by the portion of each genome that is not shared, including around 6% of functional genes that are unique to either humans or chimps. [115]

In other words, the considerable observable differences between humans and chimps may be due as much or more to genome level variation in the number, function and expression of genes rather than DNA sequence changes in shared genes. Indeed, even within humans, there has been found to be a previously unappreciated amount of copy number variation (CNV) which can make up as much as 5 – 15% of the human genome. In other words, between humans, there could be +/- 500,000,000 base pairs of DNA, some being active genes, others inactivated, or active at different levels. The full significance of this finding remains to be seen. On average, a typical human protein-coding gene differs from its chimpanzee ortholog by only two amino acid substitutions nearly one third of human genes have exactly the same protein translation as their chimpanzee orthologs. A major difference between the two genomes is human chromosome 2, which is equivalent to a fusion product of chimpanzee chromosomes 12 and 13. [116] (later renamed to chromosomes 2A and 2B, respectively).

Humans have undergone an extraordinary loss of olfactory receptor genes during our recent evolution, which explains our relatively crude sense of smell compared to most other mammals. Evolutionary evidence suggests that the emergence of color vision in humans and several other primate species has diminished the need for the sense of smell. [117]

In September 2016, scientists reported that, based on human DNA genetic studies, all non-Africans in the world today can be traced to a single population that exited Africa between 50,000 and 80,000 years ago. [118]

The human mitochondrial DNA is of tremendous interest to geneticists, since it undoubtedly plays a role in mitochondrial disease. It also sheds light on human evolution for example, analysis of variation in the human mitochondrial genome has led to the postulation of a recent common ancestor for all humans on the maternal line of descent (see Mitochondrial Eve).

Due to the lack of a system for checking for copying errors, [119] mitochondrial DNA (mtDNA) has a more rapid rate of variation than nuclear DNA. This 20-fold higher mutation rate allows mtDNA to be used for more accurate tracing of maternal ancestry. [ citation needed ] Studies of mtDNA in populations have allowed ancient migration paths to be traced, such as the migration of Native Americans from Siberia [120] or Polynesians from southeastern Asia. [ citation needed ] It has also been used to show that there is no trace of Neanderthal DNA in the European gene mixture inherited through purely maternal lineage. [121] Due to the restrictive all or none manner of mtDNA inheritance, this result (no trace of Neanderthal mtDNA) would be likely unless there were a large percentage of Neanderthal ancestry, or there was strong positive selection for that mtDNA. For example, going back 5 generations, only 1 of a person's 32 ancestors contributed to that person's mtDNA, so if one of these 32 was pure Neanderthal an expected

3% of that person's autosomal DNA would be of Neanderthal origin, yet they would have a

97% chance of having no trace of Neanderthal mtDNA. [ citation needed ]

Epigenetics describes a variety of features of the human genome that transcend its primary DNA sequence, such as chromatin packaging, histone modifications and DNA methylation, and which are important in regulating gene expression, genome replication and other cellular processes. Epigenetic markers strengthen and weaken transcription of certain genes but do not affect the actual sequence of DNA nucleotides. DNA methylation is a major form of epigenetic control over gene expression and one of the most highly studied topics in epigenetics. During development, the human DNA methylation profile experiences dramatic changes. In early germ line cells, the genome has very low methylation levels. These low levels generally describe active genes. As development progresses, parental imprinting tags lead to increased methylation activity. [122] [123]

Epigenetic patterns can be identified between tissues within an individual as well as between individuals themselves. Identical genes that have differences only in their epigenetic state are called epialleles. Epialleles can be placed into three categories: those directly determined by an individual's genotype, those influenced by genotype, and those entirely independent of genotype. The epigenome is also influenced significantly by environmental factors. Diet, toxins, and hormones impact the epigenetic state. Studies in dietary manipulation have demonstrated that methyl-deficient diets are associated with hypomethylation of the epigenome. Such studies establish epigenetics as an important interface between the environment and the genome. [124]

  1. ^"GRCh38.p13". ncbi. Genome Reference Consortium . Retrieved 8 June 2020 .
  2. ^
  3. Brown TA (2002). The Human Genome (2nd ed.). Oxford: Wiley-Liss.
  4. ^ ab
  5. Abecasis GR, Auton A, Brooks LD, DePristo MA, Durbin RM, Handsaker RE, Kang HM, Marth GT, McVean GA (November 2012). "An integrated map of genetic variation from 1,092 human genomes". Nature. 491 (7422): 56–65. Bibcode:2012Natur.491. 56T. doi:10.1038/nature11632. PMC3498066 . PMID23128226.
  6. ^
  7. Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO, et al. (October 2015). "A global reference for human genetic variation". Nature. 526 (7571): 68–74. Bibcode:2015Natur.526. 68T. doi:10.1038/nature15393. PMC4750478 . PMID26432245.
  8. ^
  9. Chimpanzee Sequencing Analysis Consortium (2005). "Initial sequence of the chimpanzee genome and comparison with the human genome" (PDF) . Nature. 437 (7055): 69–87. Bibcode:2005Natur.437. 69.. doi: 10.1038/nature04072 . PMID16136131. S2CID2638825.
  10. ^
  11. Varki A, Altheide TK (December 2005). "Comparing the human and chimpanzee genomes: searching for needles in a haystack". Genome Research. 15 (12): 1746–58. doi: 10.1101/gr.3737405 . PMID16339373.
  12. ^
  13. Wade N (23 September 1999). "Number of Human Genes Is Put at 140,000, a Significant Gain". The New York Times.
  14. ^ ab
  15. International Human Genome Sequencing Consortium (October 2004). "Finishing the euchromatic sequence of the human genome". Nature. 431 (7011): 931–45. Bibcode:2004Natur.431..931H. doi: 10.1038/nature03001 . PMID15496913.
  16. ^
  17. Ezkurdia I, Juan D, Rodriguez JM, Frankish A, Diekhans M, Harrow J, Vazquez J, Valencia A, Tress ML (November 2014). "Multiple evidence strands suggest that there may be as few as 19,000 human protein-coding genes". Human Molecular Genetics. 23 (22): 5866–78. doi:10.1093/hmg/ddu309. PMC4204768 . PMID24939910.
  18. ^
  19. Saey TH (17 September 2018). "A recount of human genes ups the number to at least 46,831". Science News.
  20. ^
  21. Alles J, Fehlmann T, Fischer U, Backes C, Galata V, Minet M, et al. (April 2019). "An estimate of the total number of true human miRNAs". Nucleic Acids Research. 47 (7): 3353–3364. doi:10.1093/nar/gkz097. PMC6468295 . PMID30820533.
  22. ^ abc
  23. Pennisi E (September 2012). "Genomics. ENCODE project writes eulogy for junk DNA". Science. 337 (6099): 1159–1161. doi:10.1126/science.337.6099.1159. PMID22955811.
  24. ^
  25. Zhang S (28 November 2018). "300 Million Letters of DNA Are Missing From the Human Genome". The Atlantic.
  26. ^ abc
  27. International Human Genome Sequencing Consortium (February 2001). "Initial sequencing and analysis of the human genome". Nature. 409 (6822): 860–921. Bibcode:2001Natur.409..860L. doi: 10.1038/35057062 . PMID11237011.
  28. ^International Human Genome Sequencing Consortium Publishes Sequence and Analysis of the Human Genome
  29. ^
  30. Pennisi E (February 2001). "The human genome". Science. 291 (5507): 1177–80. doi:10.1126/science.291.5507.1177. PMID11233420. S2CID38355565.
  31. ^
  32. Molteni M (19 November 2018). "Now You Can Sequence Your Whole Genome For Just $200". Wired.
  33. ^
  34. Wrighton K (February 2021). "Filling in the gaps telomere to telomere". Nature Milestones: Genomic Sequencing: S21.
  35. ^
  36. Pollack A (2 June 2016). "Scientists Announce HGP-Write, Project to Synthesize the Human Genome". New York Times . Retrieved 2 June 2016 .
  37. ^
  38. Boeke JD, Church G, Hessel A, Kelley NJ, Arkin A, Cai Y, et al. (July 2016). "The Genome Project-Write". Science. 353 (6295): 126–7. Bibcode:2016Sci. 353..126B. doi:10.1126/science.aaf6850. PMID27256881. S2CID206649424.
  39. ^
  40. Zhang S (28 November 2018). "300 Million Letters of DNA Are Missing From the Human Genome". The Atlantic . Retrieved 16 August 2019 .
  41. ^
  42. Chaisson MJ, Huddleston J, Dennis MY, Sudmant PH, Malig M, Hormozdiari F, et al. (January 2015). "Resolving the complexity of the human genome using single-molecule sequencing". Nature. 517 (7536): 608–11. Bibcode:2015Natur.517..608C. doi:10.1038/nature13907. PMC4317254 . PMID25383537.
  43. ^
  44. Miga KH, Koren S, Rhie A, Vollger MR, Gershman A, Bzikadze A, et al. (September 2020). "Telomere-to-telomere assembly of a complete human X chromosome". Nature. 585 (7823): 79–84. Bibcode:2020Natur.585. 79M. doi:10.1038/s41586-020-2547-7. PMC7484160 . PMID32663838.
  45. ^Ensembl genome browser release 87 [permanent dead link] (December 2016) for most values Ensembl genome browser release 68 (July 2012) for miRNA, rRNA, snRNA, snoRNA.
  46. ^
  47. Piovesan A, Pelleri MC, Antonaros F, Strippoli P, Caracausi M, Vitale L (February 2019). "On the length, weight and GC content of the human genome". BMC Research Notes. 12 (1): 106. doi:10.1186/s13104-019-4137-z. PMC6391780 . PMID30813969.
  48. ^
  49. Salzberg SL (August 2018). "Open questions: How many genes do we have?". BMC Biology. 16 (1): 94. doi:10.1186/s12915-018-0564-x. PMC6100717 . PMID30124169.
  50. ^
  51. "Gencode statistics, version 28". Archived from the original on 2 March 2018 . Retrieved 12 July 2018 .
  52. ^
  53. "Ensembl statistics for version 92.38, corresponding to Gencode v28" . Retrieved 12 July 2018 .
  54. ^
  55. "NCBI Homo sapiens Annotation Release 108". NIH. 2016.
  56. ^
  57. "CHESS statistics, version 2.0". Center for Computational Biology. Johns Hopkins University.
  58. ^
  59. "Human Genome Project Completion: Frequently Asked Questions". National Human Genome Research Institute (NHGRI) . Retrieved 2 February 2019 .
  60. ^
  61. Christley S, Lu Y, Li C, Xie X (January 2009). "Human genomes as email attachments". Bioinformatics. 25 (2): 274–5. doi: 10.1093/bioinformatics/btn582 . PMID18996942.
  62. ^
  63. Liu Z, Venkatesh SS, Maley CC (October 2008). "Sequence space coverage, entropy of genomes and the potential to detect non-human DNA in human samples". BMC Genomics. 9: 509. doi:10.1186/1471-2164-9-509. PMC2628393 . PMID18973670. , fig. 6, using the Lempel-Ziv estimators of entropy rate.
  64. ^
  65. Waters K (7 March 2007). "Molecular Genetics". Stanford Encyclopedia of Philosophy . Retrieved 18 July 2013 .
  66. ^
  67. Gannett L (26 October 2008). "The Human Genome Project". Stanford Encyclopedia of Philosophy . Retrieved 18 July 2013 .
  68. ^PANTHER Pie Chart at the PANTHER Classification System homepage. Retrieved 25 May 2011
  69. ^List of human proteins in the Uniprot Human reference proteome accessed 28 January 2015
  70. ^
  71. Kauffman SA (March 1969). "Metabolic stability and epigenesis in randomly constructed genetic nets". Journal of Theoretical Biology. 22 (3): 437–67. doi:10.1016/0022-5193(69)90015-0. PMID5803332.
  72. ^
  73. Ohno S (1972). "An argument for the genetic simplicity of man and other mammals". Journal of Human Evolution. 1 (6): 651–662. doi:10.1016/0047-2484(72)90011-5.
  74. ^
  75. Sémon M, Mouchiroud D, Duret L (February 2005). "Relationship between gene expression and GC-content in mammals: statistical significance and biological relevance". Human Molecular Genetics. 14 (3): 421–7. doi: 10.1093/hmg/ddi038 . PMID15590696.
  76. ^ M. Huang, H. Zhu, B. Shen, G. Gao, "A non-random gait through the human genome", 3rd International Conference on Bioinformatics and Biomedical Engineering (UCBBE, 2009), 1–3
  77. ^ ab
  78. Bang ML, Centner T, Fornoff F, Geach AJ, Gotthardt M, McNabb M, Witt CC, Labeit D, Gregorio CC, Granzier H, Labeit S (2001). "The complete gene sequence of titin, expression of an unusual approximately 700-kDa titin isoform, and its interaction with obscurin identify a novel Z-line to I-band linking system". Circulation Research. 89 (11): 1065–72. doi: 10.1161/hh2301.100981 . PMID11717165.
  79. ^ ab
  80. Piovesan A, Caracausi M, Antonaros F, Pelleri MC, Vitale L (2016). "GeneBase 1.1: a tool to summarize data from NCBI gene datasets and its application to an update of human gene statistics". Database: The Journal of Biological Databases and Curation. 2016: baw153. doi:10.1093/database/baw153. PMC5199132 . PMID28025344.
  81. ^Ensembl genome browser (July 2012)
  82. ^
  83. Gregory TR (September 2005). "Synergy between sequence and size in large-scale genomics". Nature Reviews Genetics. 6 (9): 699–708. doi:10.1038/nrg1674. PMID16151375. S2CID24237594.
  84. ^ ab
  85. Palazzo AF, Akef A (June 2012). "Nuclear export as a key arbiter of "mRNA identity" in eukaryotes". Biochimica et Biophysica Acta (BBA) - Gene Regulatory Mechanisms. 1819 (6): 566–77. doi:10.1016/j.bbagrm.2011.12.012. PMID22248619.
  86. ^
  87. Ludwig MZ (December 2002). "Functional evolution of noncoding DNA". Current Opinion in Genetics & Development. 12 (6): 634–9. doi:10.1016/S0959-437X(02)00355-6. PMID12433575.
  88. ^
  89. Martens JA, Laprade L, Winston F (June 2004). "Intergenic transcription is required to repress the Saccharomyces cerevisiae SER3 gene". Nature. 429 (6991): 571–4. Bibcode:2004Natur.429..571M. doi:10.1038/nature02538. PMID15175754. S2CID809550.
  90. ^
  91. Tsai MC, Manor O, Wan Y, Mosammaparast N, Wang JK, Lan F, Shi Y, Segal E, Chang HY (August 2010). "Long noncoding RNA as modular scaffold of histone modification complexes". Science. 329 (5992): 689–93. Bibcode:2010Sci. 329..689T. doi:10.1126/science.1192002. PMC2967777 . PMID20616235.
  92. ^
  93. Bartolomei MS, Zemel S, Tilghman SM (May 1991). "Parental imprinting of the mouse H19 gene". Nature. 351 (6322): 153–5. Bibcode:1991Natur.351..153B. doi:10.1038/351153a0. PMID1709450. S2CID4364975.
  94. ^
  95. Kobayashi T, Ganley AR (September 2005). "Recombination regulation by transcription-induced cohesin dissociation in rDNA repeats". Science. 309 (5740): 1581–4. Bibcode:2005Sci. 309.1581K. doi:10.1126/science.1116102. PMID16141077. S2CID21547462.
  96. ^
  97. Salmena L, Poliseno L, Tay Y, Kats L, Pandolfi PP (August 2011). "A ceRNA hypothesis: the Rosetta Stone of a hidden RNA language?". Cell. 146 (3): 353–8. doi:10.1016/j.cell.2011.07.014. PMC3235919 . PMID21802130.
  98. ^
  99. Pei B, Sisu C, Frankish A, Howald C, Habegger L, Mu XJ, Harte R, Balasubramanian S, Tanzer A, Diekhans M, Reymond A, Hubbard TJ, Harrow J, Gerstein MB (2012). "The GENCODE pseudogene resource". Genome Biology. 13 (9): R51. doi:10.1186/gb-2012-13-9-r51. PMC3491395 . PMID22951037.
  100. ^
  101. Gilad Y, Man O, Pääbo S, Lancet D (March 2003). "Human specific loss of olfactory receptor genes". Proceedings of the National Academy of Sciences of the United States of America. 100 (6): 3324–7. Bibcode:2003PNAS..100.3324G. doi:10.1073/pnas.0535697100. PMC152291 . PMID12612342.
  102. ^
  103. Iyer MK, Niknafs YS, Malik R, Singhal U, Sahu A, Hosono Y, Barrette TR, Prensner JR, Evans JR, Zhao S, Poliakov A, Cao X, Dhanasekaran SM, Wu YM, Robinson DR, Beer DG, Feng FY, Iyer HK, Chinnaiyan AM (March 2015). "The landscape of long noncoding RNAs in the human transcriptome". Nature Genetics. 47 (3): 199–208. doi:10.1038/ng.3192. PMC4417758 . PMID25599403.
  104. ^
  105. Eddy SR (December 2001). "Non-coding RNA genes and the modern RNA world". Nature Reviews Genetics. 2 (12): 919–29. doi:10.1038/35103511. PMID11733745. S2CID18347629.
  106. ^
  107. Managadze D, Lobkovsky AE, Wolf YI, Shabalina SA, Rogozin IB, Koonin EV (2013). "The vast, conserved mammalian lincRNome". PLOS Computational Biology. 9 (2): e1002917. Bibcode:2013PLSCB. 9E2917M. doi:10.1371/journal.pcbi.1002917. PMC3585383 . PMID23468607.
  108. ^
  109. Palazzo AF, Lee ES (2015). "Non-coding RNA: what is functional and what is junk?". Frontiers in Genetics. 6: 2. doi:10.3389/fgene.2015.00002. PMC4306305 . PMID25674102.
  110. ^
  111. Mattick JS, Makunin IV (April 2006). "Non-coding RNA". Human Molecular Genetics. 15 Spec No 1: R17–29. doi: 10.1093/hmg/ddl046 . PMID16651366.
  112. ^ ab
  113. Bernstein BE, Birney E, Dunham I, Green ED, Gunter C, Snyder M (September 2012). "An integrated encyclopedia of DNA elements in the human genome". Nature. 489 (7414): 57–74. Bibcode:2012Natur.489. 57T. doi:10.1038/nature11247. PMC3439153 . PMID22955616.
  114. ^
  115. Birney E (5 September 2012). "ENCODE: My own thoughts". Ewan's Blog: Bioinformatician at large.
  116. ^
  117. Stamatoyannopoulos JA (September 2012). "What does our genome encode?". Genome Research. 22 (9): 1602–11. doi:10.1101/gr.146506.112. PMC3431477 . PMID22955972.
  118. ^
  119. Carroll SB, Gompel N, Prudhomme B (May 2008). "Regulating Evolution". Scientific American. 298 (5): 60–67. Bibcode:2008SciAm.298e..60C. doi:10.1038/scientificamerican0508-60. PMID18444326.
  120. ^
  121. Miller JH, Ippen K, Scaife JG, Beckwith JR (1968). "The promoter-operator region of the lac operon of Escherichia coli". J. Mol. Biol. 38 (3): 413–20. doi:10.1016/0022-2836(68)90395-1. PMID4887877.
  122. ^
  123. Wright S, Rosenthal A, Flavell R, Grosveld F (1984). "DNA sequences required for regulated expression of beta-globin genes in murine erythroleukemia cells". Cell. 38 (1): 265–73. doi:10.1016/0092-8674(84)90548-8. PMID6088069. S2CID34587386.
  124. ^
  125. Nei M, Xu P, Glazko G (February 2001). "Estimation of divergence times from multiprotein sequences for a few mammalian species and several distantly related organisms". Proceedings of the National Academy of Sciences of the United States of America. 98 (5): 2497–502. Bibcode:2001PNAS. 98.2497N. doi:10.1073/pnas.051611498. PMC30166 . PMID11226267.
  126. ^
  127. Loots GG, Locksley RM, Blankespoor CM, Wang ZE, Miller W, Rubin EM, Frazer KA (April 2000). "Identification of a coordinate regulator of interleukins 4, 13, and 5 by cross-species sequence comparisons". Science. 288 (5463): 136–40. Bibcode:2000Sci. 288..136L. doi:10.1126/science.288.5463.136. PMID10753117. Summary
  128. ^
  129. Meunier M. "Genoscope and Whitehead announce a high sequence coverage of the Tetraodon nigroviridis genome". Genoscope. Archived from the original on 16 October 2006 . Retrieved 12 September 2006 .
  130. ^
  131. Romero IG, Ruvinsky I, Gilad Y (July 2012). "Comparative studies of gene expression and the evolution of gene regulation". Nature Reviews Genetics. 13 (7): 505–16. doi:10.1038/nrg3229. PMC4034676 . PMID22705669.
  132. ^
  133. Schmidt D, Wilson MD, Ballester B, Schwalie PC, Brown GD, Marshall A, Kutter C, Watt S, Martinez-Jimenez CP, Mackay S, Talianidis I, Flicek P, Odom DT (May 2010). "Five-vertebrate ChIP-seq reveals the evolutionary dynamics of transcription factor binding". Science. 328 (5981): 1036–40. Bibcode:2010Sci. 328.1036S. doi:10.1126/science.1186176. PMC3008766 . PMID20378774.
  134. ^
  135. Wilson MD, Barbosa-Morais NL, Schmidt D, Conboy CM, Vanes L, Tybulewicz VL, Fisher EM, Tavaré S, Odom DT (October 2008). "Species-specific transcription in mice carrying human chromosome 21". Science. 322 (5900): 434–8. Bibcode:2008Sci. 322..434W. doi:10.1126/science.1160930. PMC3717767 . PMID18787134.
  136. ^
  137. Treangen TJ, Salzberg SL (January 2012). "Repetitive DNA and next-generation sequencing: computational challenges and solutions". Nature Reviews Genetics. 13 (1): 36–46. doi:10.1038/nrg3117. PMC3324860 . PMID22124482.
  138. ^
  139. Duitama J, Zablotskaya A, Gemayel R, Jansen A, Belet S, Vermeesch JR, Verstrepen KJ, Froyen G (May 2014). "Large-scale analysis of tandem repeat variability in the human genome". Nucleic Acids Research. 42 (9): 5728–41. doi:10.1093/nar/gku212. PMC4027155 . PMID24682812.
  140. ^
  141. Pierce BA (2012). Genetics : a conceptual approach (4th ed.). New York: W.H. Freeman. pp. 538–540. ISBN978-1-4292-3250-0 .
  142. ^
  143. Bennett EA, Keller H, Mills RE, Schmidt S, Moran JV, Weichenrieder O, Devine SE (December 2008). "Active Alu retrotransposons in the human genome". Genome Research. 18 (12): 1875–83. doi:10.1101/gr.081737.108. PMC2593586 . PMID18836035.
  144. ^
  145. Liang KH, Yeh CT (2013). "A gene expression restriction network mediated by sense and antisense Alu sequences located on protein-coding messenger RNAs". BMC Genomics. 14: 325. doi:10.1186/1471-2164-14-325. PMC3655826 . PMID23663499.
  146. ^
  147. Brouha B, Schustak J, Badge RM, Lutz-Prigge S, Farley AH, Moran JV, Kazazian HH (April 2003). "Hot L1s account for the bulk of retrotransposition in the human population". Proceedings of the National Academy of Sciences of the United States of America. 100 (9): 5280–5. Bibcode:2003PNAS..100.5280B. doi:10.1073/pnas.0831042100. PMC154336 . PMID12682288.
  148. ^
  149. Barton NH, Briggs DE, Eisen JA, Goldstein DB, Patel NH (2007). Evolution. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press. ISBN978-0-87969-684-9 .
  150. ^
  151. NCBI. "GRCh38 – hg38 – Genome – Assembly – NCBI". . Retrieved 15 March 2019 .
  152. ^
  153. "from Bill Clinton's 2000 State of the Union address". Archived from the original on 21 February 2017 . Retrieved 14 June 2007 .
  154. ^
  155. Redon R, Ishikawa S, Fitch KR, Feuk L, Perry GH, Andrews TD, et al. (November 2006). "Global variation in copy number in the human genome". Nature. 444 (7118): 444–54. Bibcode:2006Natur.444..444R. doi:10.1038/nature05329. PMC2669898 . PMID17122850.
  156. ^
  157. "What's a Genome?". 15 January 2003 . Retrieved 31 May 2009 .
  158. ^
  159. NCBI_user_services (29 March 2004). "Mapping Factsheet". Archived from the original on 19 July 2010 . Retrieved 31 May 2009 .
  160. ^
  161. "About the Project". HapMap . Retrieved 31 May 2009 .
  162. ^
  163. "2008 Release: Researchers Produce First Sequence Map of Large-Scale Structural Variation in the Human Genome". . Retrieved 31 May 2009 .
  164. ^
  165. Kidd JM, Cooper GM, Donahue WF, Hayden HS, Sampas N, Graves T, et al. (May 2008). "Mapping and sequencing of structural variation from eight human genomes". Nature. 453 (7191): 56–64. Bibcode:2008Natur.453. 56K. doi:10.1038/nature06862. PMC2424287 . PMID18451855.
  166. ^ ab
  167. Abel HJ, Larson DE, Regier AA, Chiang C, Das I, Kanchi KL, et al. (July 2020). "Mapping and characterization of structural variation in 17,795 human genomes". Nature. 583 (7814): 83–89. doi:10.1038/s41586-020-2371-0. PMC7547914 . PMID32460305.
  168. ^
  169. Gray IC, Campbell DA, Spurr NK (2000). "Single nucleotide polymorphisms as tools in human genetics". Human Molecular Genetics. 9 (16): 2403–2408. doi: 10.1093/hmg/9.16.2403 . PMID11005795.
  170. ^
  171. Lai E (June 2001). "Application of SNP technologies in medicine: lessons learned and future challenges". Genome Research. 11 (6): 927–9. doi: 10.1101/gr.192301 . PMID11381021.
  172. ^
  173. "Human Genome Project Completion: Frequently Asked Questions". . Retrieved 31 May 2009 .
  174. ^
  175. Singer E (4 September 2007). "Craig Venter's Genome". MIT Technology Review . Retrieved 25 May 2010 .
  176. ^
  177. Pushkarev D, Neff NF, Quake SR (September 2009). "Single-molecule sequencing of an individual human genome". Nature Biotechnology. 27 (9): 847–50. doi:10.1038/nbt.1561. PMC4117198 . PMID19668243.
  178. ^
  179. Ashley EA, Butte AJ, Wheeler MT, Chen R, Klein TE, Dewey FE, et al. (May 2010). "Clinical assessment incorporating a personal genome". Lancet. 375 (9725): 1525–35. doi:10.1016/S0140-6736(10)60452-7. PMC2937184 . PMID20435227.
  180. ^
  181. Dewey FE, Chen R, Cordero SP, Ormond KE, Caleshu C, Karczewski KJ, et al. (September 2011). "Phased whole-genome genetic risk in a family quartet using a major allele reference sequence". PLOS Genetics. 7 (9): e1002280. doi: 10.1371/journal.pgen.1002280 . PMC3174201 . PMID21935354.
  182. ^
  183. "Complete Genomics Adds 29 High-Coverage, Complete Human Genome Sequencing Datasets to Its Public Genomic Repository".
  184. ^
  185. Sample I (17 February 2010). "Desmond Tutu's genome sequenced as part of genetic diversity study". The Guardian.
  186. ^
  187. Schuster SC, Miller W, Ratan A, Tomsho LP, Giardine B, Kasson LR, et al. (February 2010). "Complete Khoisan and Bantu genomes from southern Africa". Nature. 463 (7283): 943–7. Bibcode:2010Natur.463..943S. doi:10.1038/nature08795. PMC3890430 . PMID20164927.
  188. ^
  189. Rasmussen M, Li Y, Lindgreen S, Pedersen JS, Albrechtsen A, Moltke I, et al. (February 2010). "Ancient human genome sequence of an extinct Palaeo-Eskimo". Nature. 463 (7282): 757–62. Bibcode:2010Natur.463..757R. doi:10.1038/nature08835. PMC3951495 . PMID20148029.
  190. ^
  191. Corpas M, Cariaso M, Coletta A, Weiss D, Harrison AP, Moran F, Yang H (12 November 2013). "A Complete Public Domain Family Genomics Dataset". bioRxiv10.1101/000216 .
  192. ^
  193. Corpas M (June 2013). "Crowdsourcing the corpasome". Source Code for Biology and Medicine. 8 (1): 13. doi:10.1186/1751-0473-8-13. PMC3706263 . PMID23799911.
  194. ^
  195. Mao Q, Ciotlos S, Zhang RY, Ball MP, Chin R, Carnevali P, et al. (October 2016). "The whole genome sequences and experimentally phased haplotypes of over 100 personal genomes". GigaScience. 5 (1): 42. doi:10.1186/s13742-016-0148-z. PMC5057367 . PMID27724973.
  196. ^
  197. Cai B, Li B, Kiga N, Thusberg J, Bergquist T, Chen YC, et al. (September 2017). "Matching phenotypes to whole genomes: Lessons learned from four iterations of the personal genome project community challenges". Human Mutation. 38 (9): 1266–1276. doi:10.1002/humu.23265. PMC5645203 . PMID28544481.
  198. ^
  199. Gonzaga-Jauregui C, Lupski JR, Gibbs RA (2012). "Human genome sequencing in health and disease". Annual Review of Medicine. 63: 35–61. doi:10.1146/annurev-med-051010-162644. PMC3656720 . PMID22248320.
  200. ^
  201. Choi M, Scholl UI, Ji W, Liu T, Tikhonova IR, Zumbo P, Nayir A, Bakkaloğlu A, Ozen S, Sanjad S, Nelson-Williams C, Farhi A, Mane S, Lifton RP (November 2009). "Genetic diagnosis by whole exome capture and massively parallel DNA sequencing". Proceedings of the National Academy of Sciences of the United States of America. 106 (45): 19096–101. Bibcode:2009PNAS..10619096C. doi:10.1073/pnas.0910672106. PMC2768590 . PMID19861545.
  202. ^ ab
  203. Narasimhan VM, Xue Y, Tyler-Smith C (April 2016). "Human Knockout Carriers: Dead, Diseased, Healthy, or Improved?". Trends in Molecular Medicine. 22 (4): 341–351. doi:10.1016/j.molmed.2016.02.006. PMC4826344 . PMID26988438.
  204. ^
  205. Saleheen D, Natarajan P, Armean IM, Zhao W, Rasheed A, Khetarpal SA, et al. (April 2017). "Human knockouts and phenotypic analysis in a cohort with a high rate of consanguinity". Nature. 544 (7649): 235–239. Bibcode:2017Natur.544..235S. doi:10.1038/nature22034. PMC5600291 . PMID28406212.
  206. ^ ab
  207. Hamosh A, Scott AF, Amberger J, Bocchini C, Valle D, McKusick VA (January 2002). "Online Mendelian Inheritance in Man (OMIM), a knowledgebase of human genes and genetic disorders". Nucleic Acids Research. 30 (1): 52–5. doi: 10.1093/nar/30.1.52 . PMC99152 . PMID11752252.
  208. ^
  209. Katsanis N (November 2016). "The continuum of causality in human genetic disorders". Genome Biology. 17 (1): 233. doi:10.1186/s13059-016-1107-9. PMC5114767 . PMID27855690.
  210. ^
  211. Wong LC (2017). "Overview of the Clinical Utility of Next Generation Sequencing in Molecular Diagnoses of Human Genetic Disorders". In Wong LC (ed.). Next Generation Sequencing Based Clinical Molecular Diagnosis of Human Genetic Disorders. Springer International Publishing. pp. 1–11. doi:10.1007/978-3-319-56418-0_1. ISBN978-3-319-56418-0 . Missing or empty |title= (help)
  212. ^
  213. Fedick A, Zhang J (2017). "Next Generation of Carrier Screening". In Wong LC (ed.). Next Generation Sequencing Based Clinical Molecular Diagnosis of Human Genetic Disorders. Springer International Publishing. pp. 339–354. doi:10.1007/978-3-319-56418-0_16. ISBN978-3-319-56418-0 . Missing or empty |title= (help)
  214. ^
  215. Waterston RH, Lindblad-Toh K, Birney E, Rogers J, Abril JF, Agarwal P, Agarwala R, Ainscough R, Alexandersson M, et al. (December 2002). "Initial sequencing and comparative analysis of the mouse genome". Nature. 420 (6915): 520–62. Bibcode:2002Natur.420..520W. doi: 10.1038/nature01262 . PMID12466850. the proportion of small (50–100 bp) segments in the mammalian genome that is under (purifying) selection can be estimated to be about 5%. This proportion is much higher than can be explained by protein-coding sequences alone, implying that the genome contains many additional features (such as untranslated regions, regulatory elements, non-protein-coding genes, and chromosomal structural elements) under selection for biological function.
  216. ^
  217. Birney E, Stamatoyannopoulos JA, Dutta A, Guigó R, Gingeras TR, Margulies EH, et al. (June 2007). "Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project". Nature. 447 (7146): 799–816. Bibcode:2007Natur.447..799B. doi:10.1038/nature05874. PMC2212820 . PMID17571346.
  218. ^
  219. The Chimpanzee Sequencing Analysis Consortium (September 2005). "Initial sequence of the chimpanzee genome and comparison with the human genome". Nature. 437 (7055): 69–87. Bibcode:2005Natur.437. 69.. doi: 10.1038/nature04072 . PMID16136131. We calculate the genome-wide nucleotide divergence between human and chimpanzee to be 1.23%, confirming recent results from more limited studies.
  220. ^
  221. The Chimpanzee Sequencing Analysis Consortium (September 2005). "Initial sequence of the chimpanzee genome and comparison with the human genome". Nature. 437 (7055): 69–87. Bibcode:2005Natur.437. 69.. doi: 10.1038/nature04072 . PMID16136131. we estimate that polymorphism accounts for 14–22% of the observed divergence rate and thus that the fixed divergence is

120 ms 8.0% Scribunto_LuaSandboxCallback::gsub 120 ms 8.0% Scribunto_LuaSandboxCallback::callParserFunction 100 ms 6.7% dataWrapper 60 ms 4.0% Scribunto_LuaSandboxCallback::plain 40 ms 2.7% format 40 ms 2.7% 40 ms 2.7% [others] 360 ms 24.0% Number of Wikibase entities loaded: 1/400 -->


Millions of single nucleotide polymorphisms (SNPs) have been collected in the public database, dbSNP [1], and it is estimated that �% of human sequence variants are SNPs [2]. Among them, non-synonymous SNPs (nsSNPs), also known as single amino acid polymorphisms (SAPs), that lead to a single amino acid change in the protein product are most relevant to human inherited diseases [3]. Two databases, the Online Mendelian Inheritance in Man (OMIM) [4] and the Human gene mutation database (HGMD) [3], contain records of disease-causing variants and suggest that the majority of the disease-causing variants are non-synonymous changes [5]. It is estimated that there are 67,000�,000 nsSNPs in the human population [5]. Some of these nsSNPs are disease-associated, while others are functionally neutral. It is important to discriminate disease-associated nsSNPs from neutral ones for the investigation of genetic diseases.

Empirical rule-based [6], [7], [8], probabilistic models [9] and machine learning approaches [10], [11], [12], [13], [14], [15], [16], [17] were used to classify the nsSNPs. These studies made use of a variety of potential features to distinguish deleterious nsSNPs from neutral ones – mainly features derived from protein sequences [11], [12], [13] or from both protein structural and sequential information [10], [14], [15], [16], [17]. However, only a limited number of proteins have known three-dimensional structures, while the vast majority does not have their structural information available [5]. Among the above mentioned papers that mainly used the sequence information, some did not consider the sequence microenvironment [13] and some lacked a feature selection procedure [16].

The major limitation of traditional methods that are based on structural or sequential features is that they only focus on the local variation of the protein itself. Although the prediction accuracy may be high, it is hard to believe that the change of only one SAP protein could determine or cause a pathophysiological phenotype. More and more studies have shown that diseases can be caused by perturbed cellular networks [18], [19]. Including network features, therefore, should improve the prediction of deleterious SAPs.

In this paper, a new classification method was established by combining new network features and traditional sequential features of the amino acid microenvironment surrounding the SAP and using a carefully designed feature selection procedure. Each SAP was coded by 472 features, which were derived from the transformed scores of the amino acid index, position-specific scoring matrices, the structural features, betweenness and the KEGG enrichment scores of the protein neighbors in STRING [20] network. Next, feature selection and analysis methods, including the Maximum Relevance Minimum Redundancy method (mRMR) [21] and Incremental Feature Selection (IFS) [22] were used to obtain the optimal features to be used for the prediction of deleterious nsSNPs versus neutral ones. The prediction model was built using well-known Nearest Neighbor Algorithm (NNA) [23]. As a result, the optimal 263-feature set were selected, achieving a correct prediction rate of 83.27% when evaluated by Jackknife cross-validation test. The optimized prediction model with 263 features was also tested on an independent dataset, and the accuracy was still 80.00%. Network features were found to be most important for accurate prediction.

PureGenomics ® SNP Peek: COMT

The SNP Peek series brings you concise, up-to-date information on genetic variations known as Single Nucleotide Polymorphisms (SNPs), which affect a significant percentage of patients. The SNPs featured in this series are clinically relevant, nutritionally actionable and validated by published research. Featuring one SNP at a time, the series will educate readers about prevalence, important research findings, targeted nutritional supplements and monitoring. ‡

To apply this information in practice quickly and easily, visit

SNP PEEK COMT (Catechol O-Methyltransferase) Val158Met (rs4680)

COMT is an enzyme that degrades dopamine, a critical neurotransmitter that regulates cognition and behavior. 1-4 COMT also detoxifies estradiol. 5 The Val158Met SNP is among the most extensively researched genetic variations in psychiatry. 1-4, 6-14 The Val allele increases COMT enzyme activity, while the Met allele limits COMT activity to 25% of its normal function, allowing dopamine and estrogen to reach higher levels. 1-5

Figure 1. Dopamine, a stimulant neurotransmitter derived from L-tyrosine, supports mental sharpness, working memory and other aspects of cognitive performance. COMT degrades dopamine, limiting its availability and actions. COMT activity is largely determined by the Val158Met SNP.

Who is Affected?

The majority of the world’s population is homozygous for the wild-type allele (known as GG, -/- or Val/Val). An estimated 20-30% of Caucasians of European ancestry are homozygous for the Met allele (Met/Met, also known as AA or +/+). This genotype appears to be less common in Asian and African populations. 6

Clinical Relevance:
  • Stress tolerance: The Val/Val (GG or -/-) genotype is associated with better cognitive and psychological tolerance of stressful situations. Conversely, the Met allele has been associated with nervousness, worry and fear. 4
  • Cognitive performance: The Met/Met (AA or +/+) genotype is associated with better performance in memory and attention tasks than other genotypes. This advantage of the Met allele is likely a result of heightened dopaminergic neurotransmission in the prefrontal cortex. However, the effect may be diminished under stressful conditions. 4
  • Response to stimulants: Stimulant drugs reduce cognitive function in Met carriers. Conversely, they tend to improve cognition in Val carriers. 7
  • Estrogen metabolism: The Met allele may increase levels of estradiol due to decreased methylation of this hormone. 5
The Research:
  • In two clinical studies, subjects with at least one Met allele exhibited better executive function and performance on memory tasks than subjects with Val/Val genotypes. 8-9
  • The Met allele may reduce caffeine tolerance. In a cohort of 773 men, heavy coffee consumption was associated with cardiovascular events in Met allele carriers. 10
  • The COMT Val allele may exacerbate elevated homocysteine levels in MTHFR T/T (+/+) genotypes, according to a study of 780 elderly adults. 11 In patients with elevated homocysteine, it is prudent to ensure adequate folate, B6 and B12 intake. ‡
Diet & Lifestyle Recommendations ‡
  • For Val/Val (-/-) and Val/Met (-/+) genotypes: Consume adequate protein, which provides amino acid precursors of dopamine and norepinephrine. Exercise also supports daily cognitive function, alertness, energy and mood.
  • For Met/Met (+/+) genotypes: Consider relaxation techniques, psychotherapy and/or meditation to manage stress. To support restful sleep, practice sleep hygiene techniques and consider magnesium supplementation. Use caution with caffeine and other stimulants, which magnify the effects of emotional stress. Include cruciferous vegetables as part of a whole-food based diet to support estrogen metabolism. ‡
Pure Encapsulations ® Products:

For Met/Met (+/+) genotypes

  • SAMe is the methyl donor for COMT. SAMe also supports cognitive function and mood through other mechanisms. ‡
  • Magnesium (glycinate) supports healthy stress responses by virtue of its role in COMT-independent neurotransmitter systems. ‡
  • Lithium (orotate) or Lithium liquid may be beneficial in supporting mood and neurocognitive health. ‡
  • Adenosyl/Hydroxy B12 or Adenosyl/Hydroxy B12 liquid may be preferable to methylcobalamin because it tends to be less stimulating, according to unpublished case reports.
  • DIM Detox supports estrogen detoxification. ‡

For Val/Val (-/-) & Val/Met (-/+) genotypes

  • DopaPlus provides dopamine precursors (L-tyrosine and L-DOPA) and cofactors (zinc, vitamin B6 and folate) to promote the production of dopamine to support daily cognitive function and performance on mental tasks. ‡
  • Rhodiola Rosea maintains healthy adrenal catecholamine activity and may support energy levels. ‡

Because neurotransmission and hormone metabolism depend on many interacting genetic and environmental factors, not all individuals with COMT variants will show clinical manifestations or require specific support.

Product selection should consider other factors, such as nutrient status (refer to suggested monitoring below), appraisal of mental and cognitive function, and other relevant information obtained in the patient evaluation. ‡

Assessment & Monitoring:
  • Magnesium has no direct relationship to COMT, but is critical for healthy stress responses, cognitive performance and emotional wellness, irrespective of genotype. RBC magnesium reflects intracellular stores of this essential mineral. ‡
  • Estrogen methylation ratio (2-OHE1:2-OMeE1) indicates how effectively a patient methylates estrogen.
To Learn More:

The following databases provide abstracts of published studies, scholarly reviews and other types of articles with reliable, up-to-date information. To retrieve all relevant published studies on COMT Val158Met, enter the accession number (rs4680) in the search field. Full text articles are available in open-access journals only.
Google Scholar:

About PureGenomics ®

PureGenomics ® is a platform combining educational tools, protocols, core products, and E-script—our electronic prescription service—with, our dynamic, practitioner-exclusive website application. is designed to help identify common genetic variations known as Single Nucleotide Polymorphisms (SNPs) that are clinically relevant and nutritionally actionable. ‡

This unique platform makes it easy to TEST, TRANSLATE and TARGET SNPs with the right nutritional support, empowering practitioners with precision and confidence in the pursuit of optimal health for every patient. ‡

Learn how to successfully implement PureGenomics ® in your practice today!


Variable mutation rates among different tumor types and mutation subtypes

As mentioned previously (Methods), we classified all the mutations into 20 subtypes based on both mutation types and di-nucleotide sequence contexts (Additional file 2: Table S2). In the COSMIC mutation dataset, skin, stomach, bladder and colon tumors have relatively high overall mutational rates, which were consistent with a previous report [4]. Besides, we also observed high mutational rates in bone and endometrium tumors (Fig.  1b ). However, we observed highly variable mutational rates across different mutation subtypes (Kruskal-Wallis H-test, p =𠂒.22e-05). For example, in bone tumors, nonsense non-CpG C/G transversion has a mutation rate of 0.69/Mb while nonsense CpG C/G transition has a mutation rate of 14.2/Mb. Similarly, the mutational rate can vary substantially across different tumor types (Kruskal-Wallis H-test, p =𠂓.49e-40). For example, missense non-CpG C/G transition has an average rate of 6.18/Mb in skin tumors, much higher than which in brain tumors (0.61/Mb). Therefore, to identify potentially drivers that are positively selected in cancer, it is important to account for variations in mutation subtype and sequence context in different tumor types, instead of examining only variant frequencies in the population.

Identifying hotspot mutations in COSMIC

We started with all the mutations in 17 tumor types in COSMIC v71 (Fig.  2 ). Only data that were obtained via either whole exome or whole genome sequencing were used (Methods, Additional file 1: Table S1) [15]. Estimation of background mutation rates may be biased by outlier hyper-mutated samples. To avoid such bias, we calculated the mean μ and the standard deviation σ of the number of mutations in each sample, labeled the samples with numbers of mutations greater than μ +𠂒σ as hyper-mutated, and excluded them from further considerations (Additional file 1: Table S1).

Illustration of hotspot mutations definition and functional utility analysis. We used COSMIC v71 data as the input. We first selected the samples that were examined with whole genome or whole exome sequencing, and then removed the hyper-mutated samples in each tumor types. Hotspot mutations were identified in individual tumor types, and the biological utility investigations were performed through multiple aspects

Our goal was to identify hotspot mutations within genes (Methods) and to explore their potentially biological utilities under different biological contexts. The large number of samples in COSMIC made it possible to reliably estimate a background mutation rate for each gene in each tumor type and mutation subtype (Methods). We identified a hotspot mutation as the set of genomic aberrations that affect an amino acid (AA) position and occur significantly more frequently than expected from the background. In total, we identified a set of 702 putative hotspot mutations in 549 genes in 17 tumor types (Fig.  2 , Methods).

We measured the composition of different mutational subtypes in the hotspot mutations (Additional file 5: Figure S1). As expected, 510 (72.65 %) were missense and 17 (2.42 %) were nonsense, occupying a high proportion of hotspot mutations. We also identified 31 insertion (4.42 %) and 78 deletion (11.11 %) hotspots, which were largely ignored in previously studies [5, 6] and potentially offered novel candidates for driver mutation and cancer gene prediction. Besides, we examined the insertion and deletion hotspots and found that 17/31 was in-frame insertions and 17/78 was in-frame deletions. Among the remaining frame-shift insertion and deletions hotspots, more than 70 % have slightly different start positions and/or sizes. For example, the ESRP1 N512 hotspot deletion has two genomic variants chr8:95686611A/- and chr8:95686611-95686612AA/-.

We found that the hotspot-mutation-containing-genes (HMCGs) identified in our study overlapped significantly (98/546 vs 451/24405, Fisher exact test, p =𠂑.28e-53) with the 546 cancer genes reported in the Caner Gene Census (CGC). Among 24,951 available genes in COSMIC, 549 genes were identified to contain at least one hotspot, among which 98 were the CGC cancer genes. Similarly, we found that HMCGs overlapped significantly with the significantly mutated genes reported in TCGA PANCAN analysis (101/435 vs 448/24516, Fisher exact test, p =𠂖.56e-74) and in Lawrence et al. (73/221 vs 476/24630, Fisher exact test, p =𠂒.56e-65). The non-overlapping genes were detected due likely to that 1) the previous studies had different background mutation rate assumptions than our study 2) they detected large number of tumor suppressors that do not contain clear hotspot mutations 3) our study was not only able to detect hotspot mutations in known cancer genes, but also capable of detecting hotspot mutations in infrequently mutated genes, which may have previously unknown biological functionality 4) our study included mutation types (indels) that previous studies did not. The extent of overlap between HMCGs and the union of the above mentioned cancer gene sets remained highly significant when we chose various adjusted p value cutoffs to identify the hotspot mutations (Additional file 6: Figure S2), which indicated the statistical robustness of our approach.

Furthermore, we found significantly overlapped genes between our set with those predicted by other cluster-based methods such as e-Driver [6] (151/552 vs 398/24499, Fisher exact test, p =𠂓.42e-139) and OncodriveCLUST [5] (106/489 vs 443/24462, Fisher exact test, p =𠂒.31e-74). Additionally, regarding the mutational clusters, we found 213 hotspots overlapped with 1125 significant mutational clusters as identified by e-Driver (213/1125 vs 489/92822, Proportional test, p =𠂒.14e-87) and 261 hotspots overlapped with 1042 significant mutational clusters as predicted by OncodriveCLUST (261/1042 vs 441/89561, Proportional test, p =𠂔.98e-121). Non-overlapping results were found due mainly to: 1) e-Driver and OncodriveCLUST predicted clusters based mainly on missense mutations in a uniform mutational background 2) our study identified not only missense hotspot mutations but also a substantial proportion of insertion (4.42 %) and deletion (11.11 %) hotspots (Additional file 5: Figure S1) 3) our study chose a more stringent statistical significance cutoff to increase the confidence of identified hotspot mutations.

The number of hotspot mutations varied to a great extent from one tumor type to another (Additional file 7: Figure S3 and Additional file 8: Table S5). Most tumor types had 5 to 100 hotspot mutations. However, colorectal cancer had 253 hotspot mutations despite its relatively small sample size (684 samples), including a high proportion of insertion (10 %) and deletion (23 %) hotspot mutations (Fig.  3 ). In contrast, only 65 hotspot mutations were found in myeloid cancer (1344 samples). Such enrichment may reflect a higher extent of genetic heterogeneity in the initiation and progression of colorectal cancer, as has been suggested previously [26, 27] and also that colorectal cancer is predominantly driven by mutations rather than by copy number alterations [28]. In addition, we examined the numbers of hotspot mutations and the total numbers of mutations (mutation burden) in each tumor type, but did not find a clear correlation between them (Additional file 9: Figure S4).

Mutational signatures of hotspot mutations in 16 tumor types. The x-axis represents the tumor types and the y-axis represent the 8 types of sequence contexts (concatenating missense, nonsense and silent mutations). Each bar represents the percentage of specific sequence contexts under which the hotspot mutations happen. In each tumor type, the addition of the percentages of different sequence contexts might be larger than 1, because one or more types of mutations may happen on a single hotspot driver mutation site

Sequence context signature of hotspot mutations

We investigated the mutational signatures of 702 hotspot mutations under different sequence contexts across different tumor types. As shown in Fig.  3 , in 7 different tumor types (stomach, ovarian, brain, breast, skin, pancreas and kidney cancer), NoCpG_CGts was the most prevalent sequence context compared to other sequence contexts under which the hotspot mutations happened (p <𠂐.05), indicating a higher strength of positive selection on DNA sequences with NoCpG_CGts mutation. In 3 tumor types (head&neck, liver, and myeloid cancer), NoCpG_CGtv appears to be the most prevalent sequence context (p <𠂐.05). In several tumor types such as brain and ovarian cancer, although NoCpG_CGtv did not act as the predominant mutation sequence context, it represented a fairly high percentage (brain: 32 % and ovarian: 35 %). However, in some tumor types such as bladder cancer, the hotspot mutations are significantly enriched in ATtv sequence context (35 %, p =𠂑.77e-2).

In terms of the specific sequence context that hotspot mutations occur across different tumor types, although insertion is not the most prevalent sequence context within breast cancer, the percentage of insertion in breast cancer (22 %) was significantly higher than in any other tumor types (p =𠂑.14e-02), similarly, the percentage of deletion in colorectal cancer (27 %) was obviously higher than in other tumor types (p =𠂑.84e-4), so as the percentage of ATts (36 %, p =𠂕.84e-3) in colorectal and ATtv (35 %, p =𠂓.73e-3) in myeloid cancer.

These observations revealed the common genomic features such as NoCpG_CGts and NoCpG_CGtv sequence context were positively selected across various tumor types as well as distinct genomic features that occurred in individual tumor types, and highlighted the significance of investigating the hotspot mutations under different sequence contexts separately to better understand their genetic complexities and functional indications.

To gain novel functional insight of these mutations that were predicted based on statistics of mutation data, we performed a set of additional statistical tests to associate these 702 hotspot mutations with functional evidences.

Exploring the biological utilities of hotspot mutations using TCGA mRNA/protein expression data

The functional consequences of mutations may manifest in two aspects: affecting the gene expression or leading to abnormal signaling pathway activity. To address these questions, we divided the mRNA and protein expression values of a set of TCGA samples into multiple groups based on the mutational status of a specific gene in these samples: having a hotspot mutation, no hotspot mutation, or no mutations [22]. Only mutations occurring at least twice were included and Mann–Whitney U tests were used to measure the difference between different groups [23]. Among 702 hotspot mutations, we found 42 hotspot mutations resulted in significant mRNA or protein expression alterations (Additional file 8: Table S5).

It is known that TP53 contains gain of function mutations associate with increased expression of TP53 [29, 30] through down-regulation of downstream targets such as MDM2/MDM4, which suppress the expression of TP53. However, it is not well investigated whether different mutations in TP53 exhibit different functions across different cancer types. Motivated by this, we examined the association of TP53 hotspot mutations and RNA and protein expression of TP53 in different cancer types. To focus on the effect of mutations on TP53 expression, we excluded samples harboring TP53 deletions (Methods). As shown in Fig.  4a , in breast invasive carcinoma (BRCA), samples with R175, R248 and R273 missense mutations have obviously higher mRNA or protein expression levels, comparing to samples with non-hotspot mutations and with no mutation in TP53. In ovarian serous cystadenocarcinoma (OV), similar effects were observed for R248 and R273, which are associated with increases in the TP53 mRNA and protein expressions (Additional file 10: Figure S5). However, in rectum adenocarcinoma (READ), although R175 is associated with increases in TP53 RNA expressions similar to what is observed in BRCA, R248 and R273 missense mutations are not significantly associated with the TP53 mRNA or protein expression, comparing to samples with non-hotspot or no mutations in TP53 (Fig.  4a ), implicating distinct functions of R248 and R273 in different disease contexts. In addition, G108 frame-shift deletion, I195 missense and R213 nonsense mutations, which were uniquely detected as hotspot mutations in BRCA, OV and READ respectively, are associated with either reduced or enhanced TP53 expression in corresponding cancer types, suggesting the functional heterogeneity of hotspot mutations in different cancer types (Fig.  4a and Additional file 10: Figure S5).

Functional implications of hotspot mutations in RNA and protein expression. a In BRCA, tumor samples with G108 deletion hotspot mutations in TP53 exhibit lower TP53 RNA expression than those with non-hotspot mutations and without TP53 mutations. In contrast, tumor samples with missense hotspot mutations (R175, Y220, R248 and R273) in TP53 show higher TP53 RNA and protein expression. In READ, tumor samples with R175 missense mutations show higher TP53 RNA and protein expression than those with non-hotspot mutations and without TP53 mutations, while R213 nonsense mutations has the opposite effect. b In BRCA, tumor samples with H1047 missense hotspot mutations in PIK3CA show higher AKT pT308 and pS473 levels than those with no mutations in PIK3CA, while in COAD, tumor samples with E542 missense hotspot mutations in PIK3CA show higher AKT pT308 and pS473 levels than those with no mutations in PIK3CA. * indicates p <𠂐.05 and ** indicates p <𠂐.001 between samples with specified hotspot mutations and samples with non-hotspot mutations in examined gene # indicates p <𠂐.05 and ## indicates p <𠂐.001 between samples with specified hotspot mutations and samples without mutations in examined gene

Instead of altering the RNA/protein level, certain mutations may be functional via altering downstream protein activity through signaling transduction. For example, activation of PIK3CA could lead to activation of downstream targets such as AKT phosphorylation [31]. A set of PIK3CA mutations have been detected and functionally investigated in various cancer types such as BRCA and colon adenocarcinoma (COAD) [32]. We examined the association of individual PIK3CA mutations and AKT activation by comparing the phosphorylated AKT levels in samples with various PIK3CA mutations to those in samples without PIK3CA mutation. Surprisingly, in BRCA, only PIK3CA H1047 was associated with dramatically higher AKT pT308 and pS473 levels, comparing to those that did not have any PIK3CA mutations (Fig.  4b ) in COAD, only PIK3CA E542 were associated with significantly higher AKT pT308 and pS473 levels, comparing to those that did not have any PIK3CA mutations (Fig.  4b ). Notably, in both cases, PIK3CA mutations did not affect the total AKT level (data not shown), suggesting that different PIK3CA mutations in different cancer types may selectively activate AKT via signaling transduction, rather than expression regulation.

The availability of mRNA and protein expression data enable an opportunity to detailed characterize the biological consequences of different mutations in one cancer type, as well as one mutation under different cancer contexts, reiterating the rationale of distinguishing the function of individual mutations in different disease contexts.

Exploring the pharmacogenomics properties of hotspot mutations

It has been shown that cancer cells respond to specific drugs when they harbor mutations in driver genes such as BRAF and NRAS [9]. However, it is not entirely clear whether different mutations in a driver gene can trigger different drug responses. Here, we assessed the effects of individual mutations on drug responsiveness using data from the CCLE [24]. We divided cancer cell-line samples into different groups, depending on whether they contain specific hotspot, non-hotspot, or no mutations in investigated gene candidates. Only mutations occurring at least twice were included and Mann–Whitney U test was performed to measure the difference [23]. Among 702 hotspot mutations, we found 35 hotspot mutations lead to significantly altered drug sensitivities (Additional file 8: Table S5).

We first illustrated the effect of individual hotspot mutations in BRAF, KRAS and NRAS on the sensitivity of cancer cells treated by MEK inhibitors (PD-0325901 and AZD6244). As expected, cells with BRAF V600E mutations demonstrated significantly higher sensitivity to MEK inhibitors than those without BRAF mutations (data not shown). Furthermore, we found that cells with NRAS Q61 hotspot mutations demonstrated significantly higher sensitivity to MEK inhibitors than those with non-hotspot mutations and those without mutations in NRAS (Fig.  5a ). Cells with KRAS G12 hotspot mutations demonstrated significantly higher sensitivity to MEK inhibitors than those with non-hotspot mutations and those without mutations in KRAS (Fig.  5a ).

Functional implications of hotspot mutations in drug sensitivity. a Cancer cells with NRAS Q61 or KRAS G12 missense hotspot mutations exhibit higher sensitivity to MEK inhibitors (PD-0325901 and AZD6244) than those with non-hotspot mutations or without any mutations in NRAS or KRAS. b Cancer cells with MAP3K4 A1199 deletion hotspot mutations exhibit lower sensitivity to different EGFR inhibitors (Erlotinib, Lapatinib, TKI258 and AZD0530) than those with non-hotspot mutations or without any mutations in MAP3K4. * indicates p <𠂐.05 between samples with specified hotspot mutations and samples with non-hotspot mutations in examined gene # indicates p <𠂐.05 between samples with specified hotspot mutations and samples without mutations in examined gene

Epidermal growth factor (EGF) is one of the high affinity ligands of EGFR. EGF/EGFR system induces cell growth, differentiation, migration, adhesion and cell survival through various interacting signaling pathways such as MAPK pathway [33], in which MAP3K4 is an important component [34]. Clinically, EGFR inhibitors such as Erlotinib were used to repress EGFR signaling activations and suppress tumor cell growth. However, we found that cancer cell-lines with MAP3K4 A1199 deletion hotspot mutations were more resistant to all four examined EGFR inhibitors (Erlotinib, Lapatinib, TKI258 and AZD0530) in comparison to cancer cell-lines without MAP3K4 mutations (Fig.  5b ). These EGFR hotspot mutant cell-lines are also more resistant to three inhibitors (Erlotinib, Lapatinib and TKI258) in comparison to cell-lines containing non-hotspot mutations in MAP3K4 (Fig.  5b ), suggesting the unique function of MAP3K4 A1199 deletion in disrupting the MAPK pathway function and its potential biomarker utility.

These observations above support that hotspot mutations we identified may have distinct roles in mediating signaling pathways and are associated with different drug sensitivities. Therefore, it is critical to obtain accurate genomic information and interpret them in context-specific manner in order to achieve desirable outcomes in personalized cancer treatment.

Tumor type-specific hotspot mutations

We performed an analysis to assess whether a hotspot mutation in our set is highly prevalent in specific tumor types. Among all the 702 hotspots, we found that 68 were highly prevalent in one tumor type, 11 in two tumor types, 2 (KRAS G12 and PIK3CA E542) in three tumor types, and 1 (KRAS G13) in four tumor types (Additional file 11: Figure S6). Among these, 34 hotspot mutations such as CD209 R129 missense (4.0 %) in bladder cancer, MAGI1 Q421 insertion (0.8 %) and NR1H2 Q175 insertion (1.8 %) in breast cancer were not well investigated based on previous studies and are potentially novel targets (Additional file 8: Table S5).

Of the 21 hotspot mutations detected in TP53 (Fig.  6a ), 2 were found to be prevalent in multiple cancer types (R248 in bladder urothelial carcinoma (BLCA), BRCA and OV, R273 in lower grade glioma (LGG), BRCA and OV), and 9 (G108, R158, R175, I195, R213, Y220, R249, R282, E285) in one tumor type, confirming the functional diversity of TP53 hotspot mutations in different cancer types (Fig.  4a ).

Prevalence of hotspot mutations in different TCGA cancer types and their functional implications. a In TP53, hotspot mutations are differentially prevalent in different tumor types, indicating their differential functions. b In BRCA, samples with NR1H2 Q175 in-frame insertion hotspot mutations have significantly lower NR1H2 expression compared to samples with NR1H2 non-hotspot mutations. c In BRCA, sample with GATA3 P409 insertion hotspot mutations have obviously higher GATA3 compared to samples without GATA3 mutation. * indicates p <𠂐.05 between samples with specified hotspot mutations and samples with non-hotspot mutations in examined gene # indicates p <𠂐.05 between samples with specified hotspot mutations and samples without mutations in examined gene

We identified 30 hotspot mutations that were exclusively detected in only one tumor type (Additional file 12: Table S6). Included were DNMT3A R882 and NPM1 W288, which occur in 14.9 and 25.6 % of acute myeloid leukemia (LAML) patients, respectively and have been shown important in LAML oncogenesis [35]. Besides these expected hotspots, we found some potentially novel hotspots. For example, we found an in-frame insertion hotspot mutation, NR1H2 Q175 in 1.8 % of BRCA patients, further investigation using BRCA mRNA expression data showed that NR1H2 Q175 insertion is associated with reduced mRNA expression of NR1H2, comparing to NR1H2 non-hotspot mutations (Mann–Whitney U test, p =𠂒.60e-2, Fig.  6b ). Although having been reported to regulate cholesterol homeostasis and tumorigenesis of liver cancer [36], the role of NR1H2 Q175 insertion in BRCA has not been well characterized. In addition, GATA3 P409, a frame-shift insertion hotspot mutation was detected in 1.6 % of BRCA patients. BRCA samples with GATA3 P409 insertions had higher expressions of GATA3 compared to samples without GATA3 mutations based on both the BRCA mRNA expression (Mann–Whitney U test, p =𠂒.03e-2) and RRPA data (Mann–Whitney U test, p =𠂕.94e-2, Fig.  6c ). Because GATA3 has been proposed as a prognostic biomarker in breast cancer [37], the high frequency of GATA3 P409 and elevated GATA3 expression in BRCA make it a potential useful therapeutic target in clinics.

Conservation and protein-domain characteristics of the hotspot mutations

In general, functional and structural important mutations are expected to locate in highly evolutionally conserved region and domain in the protein. To evaluate our hotspot mutation, we used the RS scores computed by GERP++ [25], to measure the evolutionary constraints across different chromosomal sites (Methods). We compared the RS score difference between the sites that belong to hotspot mutations and those belong to non-hotspot mutations. The RS scores of 702 hotspot mutations were significantly higher than those of non-hotspot mutations (Fig.  7a ), suggesting the sites that harbor hotspot mutations were more conserved than those do not. In addition, we also examined the relative location of mutations on the protein. The non-hotspot mutations were evenly distributed across different domains of the protein (lower panel), while the hotspot mutations showed clustering in the middle and the terminals (Fig.  7b , upper panel), suggesting the functional preference of mutations in different protein domains.

Compare the conservation and proteomic domain localization of the hotspot and the non-hotspot mutations. a Comparison of GERP score between the hotspot and non-hotspot mutations. b Investigation of the proteomic domain location of the hotspot (upper) and non-hotspot (lower) mutations


Author contributions: N.R., J.D.J., and T.F.K. designed research N.R., J.D.J., and T.F.K. performed research L.G., M.R.S., K.H., A.Y.Y., M.M.M.-P., and W.J.B. contributed new reagents/analytic tools N.R., C.P., B.B., J.D.J., and T.F.K. analyzed data and N.R., C.P., L.G., B.B., M.R.S., K.H., A.Y.Y., M.M.M.-P., W.J.B., J.D.J., and T.F.K. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

Data deposition: The sequences reported in this paper have been deposited in the GenBank database (BioProject ID PRJNA279971).

Watch the video: The different types of mutations. Biomolecules. MCAT. Khan Academy (June 2022).


  1. Oxton

    Very interesting idea

  2. Ayub

    I am sorry, that has interfered... At me a similar situation. I invite to discussion.

  3. Erec

    I fully share your opinion. There is something in this and I think this is a great idea. I completely agree with you.

  4. Alarik

    Gorgeous, I'll take it to my diary

  5. Mikat

    Thanks for the help in this question, can, I too can help you something?

Write a message