Calibration of computational tools for missense variant pathogenicity classification and ClinGen recommendations for PP3/BP4 criteria

The American Journal of Human Genetics, Journal Year: 2022, Volume and Issue: 109(12), P. 2163 - 2177

Published: Nov. 21, 2022

Recommendations from the American College of Medical Genetics and Genomics Association for Molecular Pathology (ACMG/AMP) interpreting sequence variants specify use computational predictors as "supporting" level evidence pathogenicity or benignity using criteria PP3 BP4, respectively. However, score intervals defined by tool developers, ACMG/AMP recommendations that require consensus multiple predictors, lack quantitative support. Previously, we described a probabilistic framework quantified strengths (supporting, moderate, strong, very strong) within recommendations. We have extended this to introduce new standard converts tool's scores BP4 strengths. Our approach is based on estimating local positive predictive value can calibrate any other continuous-scale variant type. estimate thresholds (score intervals) corresponding each strength thirteen missense interpretation tools, carefully assembled independent data sets. Most tools achieved supporting both pathogenic benign classification newly established thresholds. Multiple reached justifying moderate several strong levels. One some variants. Based these findings, provide evidence-based revisions individual future assessment methods clinical interpretation.

Language: Английский

---

Big Data in Public Health: Terminology, Machine Learning, and Privacy

Annual Review of Public Health, Journal Year: 2017, Volume and Issue: 39(1), P. 95 - 112

Published: Dec. 20, 2017

The digital world is generating data at a staggering and still increasing rate. While these “big data” have unlocked novel opportunities to understand public health, they hold greater potential for research practice. This review explores several key issues that arisen around big data. First, we propose taxonomy of sources clarify terminology identify threads common across some subtypes Next, consider health practice uses data, including surveillance, hypothesis-generating research, causal inference, while exploring the role machine learning may play in each use. We then ethical implications revolution with particular emphasis on maintaining appropriate care privacy which technology rapidly changing social norms regarding need (and even meaning of) privacy. Finally, make suggestions structuring teams training succeed working

Language: Английский

---

MetaRNN: differentiating rare pathogenic and rare benign missense SNVs and InDels using deep learning

Genome Medicine, Journal Year: 2022, Volume and Issue: 14(1)

Published: Oct. 8, 2022

Multiple computational approaches have been developed to improve our understanding of genetic variants. However, their ability identify rare pathogenic variants from benign ones is still lacking. Using context annotations and deep learning methods, we present pathogenicity prediction models, MetaRNN MetaRNN-indel, help prioritize nonsynonymous single nucleotide (nsSNVs) non-frameshift insertion/deletions (nfINDELs). We use independent test sets demonstrate that these new models outperform state-of-the-art competitors achieve a more interpretable score distribution. Importantly, scores both are comparable, enabling easy adoption integrated genotype-phenotype association analysis methods. All pre-computed nsSNV available at http://www.liulab.science/MetaRNN . The stand-alone program also https://github.com/Chang-Li2019/MetaRNN

Language: Английский

---

How chromosomal inversions reorient the evolutionary process

Journal of Evolutionary Biology, Journal Year: 2023, Volume and Issue: 36(12), P. 1761 - 1782

Published: Nov. 9, 2023

Abstract Inversions are structural mutations that reverse the sequence of a chromosome segment and reduce effective rate recombination in heterozygous state. They play major role adaptation, as well other evolutionary processes such speciation. Although inversions have been studied since 1920s, they remain difficult to investigate because reduced conferred by them strengthens effects drift hitchhiking, which turn can obscure signatures selection. Nonetheless, numerous found be under Given recent advances population genetic theory empirical study, here we review how different mechanisms selection affect evolution inversions. A key difference between mutations, single nucleotide variants, is fitness an inversion may affected larger number frequently interacting processes. This considerably complicates analysis causes underlying We discuss extent these disentangled, approach. often roles adaptation speciation, but direct their obscured characteristic makes so unique (reduced arrangements). In this review, examine impact evolution, weaving together both theoretical studies. emphasize most patterns overdetermined (i.e. caused multiple processes), highlight new technologies provide path forward towards disentangling mechanisms.

Language: Английский

---

CAGI, the Critical Assessment of Genome Interpretation, establishes progress and prospects for computational genetic variant interpretation methods

Genome biology, Journal Year: 2024, Volume and Issue: 25(1)

Published: Feb. 22, 2024

Abstract Background The Critical Assessment of Genome Interpretation (CAGI) aims to advance the state-of-the-art for computational prediction genetic variant impact, particularly where relevant disease. five complete editions CAGI community experiment comprised 50 challenges, in which participants made blind predictions phenotypes from data, and these were evaluated by independent assessors. Results Performance was strong clinical pathogenic variants, including some difficult-to-diagnose cases, extends interpretation cancer-related variants. Missense methods able estimate biochemical effects with increasing accuracy. regulatory variants complex trait disease risk less definitive indicates performance potentially suitable auxiliary use clinic. Conclusions show that while current are imperfect, they have major utility research applications. Emerging increasingly large, robust datasets training assessment promise further progress ahead.

Language: Английский

---

VarSite: Disease variants and protein structure

Protein Science, Journal Year: 2019, Volume and Issue: 29(1), P. 111 - 119

Published: Oct. 13, 2019

VarSite is a web server mapping known disease-associated variants from UniProt and ClinVar, together with natural gnomAD, onto protein 3D structures in the Protein Data Bank. The analyses are primarily image-based provide both an overview for each human protein, as well report any specific variant of interest. information can be useful assessing whether given might pathogenic or benign. structural annotations position include secondary structure, interactions ligand, metal, DNA/RNA, other various measures variant's possible impact on protein's function. locations viewed interactively via 3dmol.js JavaScript viewer, RasMol PyMOL. Users search variants, sets by providing DNA coordinates base change(s) Additionally, agglomerative given, such disease Pfam CATH domains. freely accessible to all at: https://www.ebi.ac.uk/thornton-srv/databases/VarSite.

Language: Английский

---

Autism risk in offspring can be assessed through quantification of male sperm mosaicism

Nature Medicine, Journal Year: 2019, Volume and Issue: 26(1), P. 143 - 150

Published: Dec. 23, 2019

Language: Английский

---

BioAider: An efficient tool for viral genome analysis and its application in tracing SARS-CoV-2 transmission

Sustainable Cities and Society, Journal Year: 2020, Volume and Issue: 63, P. 102466 - 102466

Published: Aug. 28, 2020

Language: Английский

---

Unsupervised and semi‐supervised learning: the next frontier in machine learning for plant systems biology

The Plant Journal, Journal Year: 2022, Volume and Issue: 111(6), P. 1527 - 1538

Published: July 13, 2022

SUMMARY Advances in high‐throughput omics technologies are leading plant biology research into the era of big data. Machine learning (ML) performs an important role systems because its excellent performance and wide application analysis However, to achieve ideal performance, supervised ML algorithms require large numbers labeled samples as training In some cases, it is impossible or prohibitively expensive obtain enough data; here, paradigms unsupervised (UL) semi‐supervised (SSL) play indispensable role. this review, we first introduce basic concepts techniques, well representative UL SSL algorithms, including clustering, dimensionality reduction, self‐supervised (self‐SL), positive‐unlabeled (PU) transfer learning. We then review recent advances applications both phenotyping research. Finally, discuss limitations highlight significance challenges strategies biology.

Language: Английский

---

Insights on variant analysis in silico tools for pathogenicity prediction

Frontiers in Genetics, Journal Year: 2022, Volume and Issue: 13

Published: Nov. 29, 2022

Molecular biology is currently a fast-advancing science. Sequencing techniques are getting cheaper, but the interpretation of genetic variants requires expertise and computational power, therefore still challenge. Next-generation sequencing releases thousands to classify them, researchers propose protocols with several parameters. Here we present review in silico pathogenicity prediction tools involved variant prioritization/classification process used by some international for analysis studies evaluating their efficiency.

Language: Английский

---