A general framework for estimating the relative pathogenicity of human genetic variants
By Martin Kircher, Daniela M. Witten, Preti Jain, Brian J. O'Roak, Gregory M. Cooper, Jay Shendure. 2014
5,961 citations recorded by the source corpus.
Current methods for annotating and interpreting human genetic variation tend to exploit a single information type (for example, conservation) and/or are restricted in scope (for example, to missense changes). Here we describe Combined Annotation–Dependent Depletion (CADD), a method for objectively integrating many diverse annotations into a single measure (C score) for each variant. We implement CADD as a support vector machine trained to differentiate 14.7 million high-frequency human-derived alleles from 14.7 million simulated variants. We precompute C scores for all 8.6 billion possible human single-nucleotide variants and enable scoring of short insertions-deletions. C scores correlate with allelic diversity, annotations of functionality, pathogenicity, disease severity, experimentally measured regulatory effects and complex trait associations, and they highly rank known pathogenic variants within individual genomes. The ability of CADD to prioritize functional, deleterious and pathogenic variants across many functional categories, effect sizes and genetic architectures is unmatched by any current single-annotation method.
Subfields: Genetics, Molecular Biology, Computational Biology & Bioinformatics
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Genetics
- B. Feng
- Eric Boerwinkle
- Sharon R. Browning
- Gonçalo R. Abecasis
- Xiaoming Liu
- Brian L. Browning
- Abhishek Niroula
- Mauno Vihinen
In Molecular Biology
- Xiaohui S. Xie
- Daniel Quang
- Jay Shendure
- Jinbo Xu
- Abhishek Niroula
- Mauno Vihinen
- B. Feng
- Eric Boerwinkle
In Computational Biology & Bioinformatics
- Jinbo Xu
- Xiaohui S. Xie
- Daniel Quang
- Yanjun Qi
- Nick V. Grishin
- Alexander Rives
- Jack Lanchantin
- Jasper Snoek
Citations in the corpus
- Opportunities and obstacles for deep learning in biology and medicine (2017, methodology): “DANN [33] adopts this approach to predict the pathogenicity of genetic variants using semi-synthetic training data from Combined Annotation-Dependent Depletion [369].”
- Evolutionary context-integrated deep sequence modeling for protein engineering (2020, methodology): “Existing tools predict the conservation effects of mutations by considering each amino acid independently (e.g., PolyPhen-2 [40] and CADD [41] ) while others exploit structure information (e.g., FoldX [42] and OSPREY [4] ).”
- CADD: predicting the deleteriousness of variants throughout the human genome (2018, positive): “We previously described ‘Combined Annotation-Dependent Depletion’ or CADD, a score that ranks genetic variants, including single nucleotide variants (SNVs) and short inserts and deletions (InDels), throughout the human genome reference assembly (5).”
- Quantitative missense variant effect prediction using large-scale mutagenesis data (2017, neutral): “Envision predictions were better than random guessing (area under ROC curve [AUROC] = 0.72), but not as good as PolyPhen2, CADD, and SIFT (AUROC = 0.86, 0.85, and 0.84, respectively; Figure S10).”
- Standards and Guidelines for the Interpretation of Sequence Variants: A Joint Consensus Recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (2015, background): “Newer tools are beginning to address additional non-coding sequences.(13)”
- Language models enable zero-shot prediction of the effects of mutations on protein function (2021, methodology): “Models have been designed specifically for proteins, using feature engineering such as Envision [36] and PolyPhen-2 [37], ensemble methods such as Revel [38], MPC [39], CADD [40], and M-CAP [41], language models such as UniRep [23, 16] and ESM [12], and other representation learning approaches [42, ”
- The human gene damage index as a gene-level approach to prioritizing exome variants (2015, methodology): “We then attributed a predicted damage impact score, C, to each variant with CADD (9).”
- High-coverage whole-genome sequencing of the expanded 1000 Genomes Project cohort including 602 trios (2021, positive): “The 1kGP resources have been collectively cited over 16,000 times to date and have been utilized for foundational applications such as genotype imputation, eQTL mapping, variant pathogenicity prioritization, population history, and evolutionary genetics studies (Almeida et al., 2014; Hara et al., 20”
- The Evaluation of Tools Used to Predict the Impact of Missense Variants Is Hindered by Two Types of Circularity (2015, methodology): “As the intersection of the training data from the tool CADD [Kircher et al., 2014] and that of all other datasets is small (fewer than a hundred variants), we systematically excluded all variants overlapping with the CADD training data from all other data sets.”
- Ensembl 2015 (2014, methodology): “The VEP off-line script version has been extended to include additional data types (PubMed IDs, Exome Sequencing Project frequencies, UniProt IDs, HGNC IDs) with new plugins for popular data sets, such as FATHMM (34), dbNSFP (35) and CADD (36).”
- dbNSFP v3.0: A One-Stop Database of Functional Predictions and Annotations for Human Non-synonymous and Splice Site SNVs (2016, background): “0, the new version added 11 new prediction scores: MetaSVM and MetaLR [Dong et al., 2015], CADD [Kircher et al., 2014], VEST3 [Carter et al., 2013], PROVEAN [Choi et al., 2012], 4× fitCons scores [Gulko et al., 2015], fathmm-MKL [Shihab et al., 2015], and DANN [Quang et al., 2015]; two conservation…”
- FunSeq2: a framework for prioritizing noncoding regulatory variants in cancer (2014, methodology): “Results from CADD and GWAVA are shown in Additional file 1: Figure S6.”