Comprehensive annotation of BRCA1 and BRCA2 missense variants by functionally validated sequence-based computational prediction models

Steven Hart, Tanya Hoskin, Hermela Shimelis, Raymond M. Moore, Bingjian Feng, Abigail Thomas, Noralane Morey Lindor, Eric Polley, David E. Goldgar, Edwin Iversen, Alvaro N.A. Monteiro, Vera Jean Suman, Fergus J Couch

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Purpose: To improve methods for predicting the impact of missense variants of uncertain significance (VUS) in BRCA1 and BRCA2 on protein function. Methods: Functional data for 248 BRCA1 and 207 BRCA2 variants from assays with established high sensitivity and specificity for damaging variants were used to recalibrate 40 in silico algorithms predicting the impact of variants on protein activity. Additional random forest (RF) and naïve voting method (NVM) metapredictors for both BRCA1 and BRCA2 were developed to increase predictive accuracy. Results: Optimized thresholds for in silico prediction models significantly improved the accuracy of predicted functional effects for BRCA1 and BRCA2 variants. In addition, new BRCA1-RF and BRCA2-RF metapredictors showed area under the curve (AUC) values of 0.92 (95% confidence interval [CI]: 0.88–0.96) and 0.90 (95% CI: 0.84–0.95), respectively. Similarly, the BRCA1-NVM and BRCA2-NVM models had AUCs of 0.93 and 0.90. The RF and NVM models were used to predict the pathogenicity of all possible missense variants in BRCA1 and BRCA2. Conclusion: The recalibrated algorithms and new metapredictors significantly improved upon current models for predicting the impact of variants in cancer risk–associated domains of BRCA1 and BRCA2. Prediction of the functional impact of all possible variants in BRCA1 and BRCA2 provides important information about the clinical relevance of variants in these genes.

Original languageEnglish (US)
Pages (from-to)1-10
Number of pages10
JournalGenetics in Medicine
DOIs
StateAccepted/In press - Jun 8 2018

Fingerprint

Politics
Computer Simulation
Area Under Curve
BRCA2 Protein
Confidence Intervals
Virulence
Sensitivity and Specificity
Forests
Genes
Neoplasms
Proteins

Keywords

  • BRCA1 and BRCA2
  • Functional evaluation
  • In silico prediction
  • Metapredictor
  • VUS

ASJC Scopus subject areas

  • Genetics(clinical)

Cite this

Comprehensive annotation of BRCA1 and BRCA2 missense variants by functionally validated sequence-based computational prediction models. / Hart, Steven; Hoskin, Tanya; Shimelis, Hermela; Moore, Raymond M.; Feng, Bingjian; Thomas, Abigail; Lindor, Noralane Morey; Polley, Eric; Goldgar, David E.; Iversen, Edwin; Monteiro, Alvaro N.A.; Suman, Vera Jean; Couch, Fergus J.

In: Genetics in Medicine, 08.06.2018, p. 1-10.

Research output: Contribution to journalArticle

Hart, Steven ; Hoskin, Tanya ; Shimelis, Hermela ; Moore, Raymond M. ; Feng, Bingjian ; Thomas, Abigail ; Lindor, Noralane Morey ; Polley, Eric ; Goldgar, David E. ; Iversen, Edwin ; Monteiro, Alvaro N.A. ; Suman, Vera Jean ; Couch, Fergus J. / Comprehensive annotation of BRCA1 and BRCA2 missense variants by functionally validated sequence-based computational prediction models. In: Genetics in Medicine. 2018 ; pp. 1-10.
@article{e6c4f50fee6143bca4cfe72f8dff9be0,
title = "Comprehensive annotation of BRCA1 and BRCA2 missense variants by functionally validated sequence-based computational prediction models",
abstract = "Purpose: To improve methods for predicting the impact of missense variants of uncertain significance (VUS) in BRCA1 and BRCA2 on protein function. Methods: Functional data for 248 BRCA1 and 207 BRCA2 variants from assays with established high sensitivity and specificity for damaging variants were used to recalibrate 40 in silico algorithms predicting the impact of variants on protein activity. Additional random forest (RF) and na{\"i}ve voting method (NVM) metapredictors for both BRCA1 and BRCA2 were developed to increase predictive accuracy. Results: Optimized thresholds for in silico prediction models significantly improved the accuracy of predicted functional effects for BRCA1 and BRCA2 variants. In addition, new BRCA1-RF and BRCA2-RF metapredictors showed area under the curve (AUC) values of 0.92 (95{\%} confidence interval [CI]: 0.88–0.96) and 0.90 (95{\%} CI: 0.84–0.95), respectively. Similarly, the BRCA1-NVM and BRCA2-NVM models had AUCs of 0.93 and 0.90. The RF and NVM models were used to predict the pathogenicity of all possible missense variants in BRCA1 and BRCA2. Conclusion: The recalibrated algorithms and new metapredictors significantly improved upon current models for predicting the impact of variants in cancer risk–associated domains of BRCA1 and BRCA2. Prediction of the functional impact of all possible variants in BRCA1 and BRCA2 provides important information about the clinical relevance of variants in these genes.",
keywords = "BRCA1 and BRCA2, Functional evaluation, In silico prediction, Metapredictor, VUS",
author = "Steven Hart and Tanya Hoskin and Hermela Shimelis and Moore, {Raymond M.} and Bingjian Feng and Abigail Thomas and Lindor, {Noralane Morey} and Eric Polley and Goldgar, {David E.} and Edwin Iversen and Monteiro, {Alvaro N.A.} and Suman, {Vera Jean} and Couch, {Fergus J}",
year = "2018",
month = "6",
day = "8",
doi = "10.1038/s41436-018-0018-4",
language = "English (US)",
pages = "1--10",
journal = "Genetics in Medicine",
issn = "1098-3600",
publisher = "Lippincott Williams and Wilkins",

}

TY - JOUR

T1 - Comprehensive annotation of BRCA1 and BRCA2 missense variants by functionally validated sequence-based computational prediction models

AU - Hart, Steven

AU - Hoskin, Tanya

AU - Shimelis, Hermela

AU - Moore, Raymond M.

AU - Feng, Bingjian

AU - Thomas, Abigail

AU - Lindor, Noralane Morey

AU - Polley, Eric

AU - Goldgar, David E.

AU - Iversen, Edwin

AU - Monteiro, Alvaro N.A.

AU - Suman, Vera Jean

AU - Couch, Fergus J

PY - 2018/6/8

Y1 - 2018/6/8

N2 - Purpose: To improve methods for predicting the impact of missense variants of uncertain significance (VUS) in BRCA1 and BRCA2 on protein function. Methods: Functional data for 248 BRCA1 and 207 BRCA2 variants from assays with established high sensitivity and specificity for damaging variants were used to recalibrate 40 in silico algorithms predicting the impact of variants on protein activity. Additional random forest (RF) and naïve voting method (NVM) metapredictors for both BRCA1 and BRCA2 were developed to increase predictive accuracy. Results: Optimized thresholds for in silico prediction models significantly improved the accuracy of predicted functional effects for BRCA1 and BRCA2 variants. In addition, new BRCA1-RF and BRCA2-RF metapredictors showed area under the curve (AUC) values of 0.92 (95% confidence interval [CI]: 0.88–0.96) and 0.90 (95% CI: 0.84–0.95), respectively. Similarly, the BRCA1-NVM and BRCA2-NVM models had AUCs of 0.93 and 0.90. The RF and NVM models were used to predict the pathogenicity of all possible missense variants in BRCA1 and BRCA2. Conclusion: The recalibrated algorithms and new metapredictors significantly improved upon current models for predicting the impact of variants in cancer risk–associated domains of BRCA1 and BRCA2. Prediction of the functional impact of all possible variants in BRCA1 and BRCA2 provides important information about the clinical relevance of variants in these genes.

AB - Purpose: To improve methods for predicting the impact of missense variants of uncertain significance (VUS) in BRCA1 and BRCA2 on protein function. Methods: Functional data for 248 BRCA1 and 207 BRCA2 variants from assays with established high sensitivity and specificity for damaging variants were used to recalibrate 40 in silico algorithms predicting the impact of variants on protein activity. Additional random forest (RF) and naïve voting method (NVM) metapredictors for both BRCA1 and BRCA2 were developed to increase predictive accuracy. Results: Optimized thresholds for in silico prediction models significantly improved the accuracy of predicted functional effects for BRCA1 and BRCA2 variants. In addition, new BRCA1-RF and BRCA2-RF metapredictors showed area under the curve (AUC) values of 0.92 (95% confidence interval [CI]: 0.88–0.96) and 0.90 (95% CI: 0.84–0.95), respectively. Similarly, the BRCA1-NVM and BRCA2-NVM models had AUCs of 0.93 and 0.90. The RF and NVM models were used to predict the pathogenicity of all possible missense variants in BRCA1 and BRCA2. Conclusion: The recalibrated algorithms and new metapredictors significantly improved upon current models for predicting the impact of variants in cancer risk–associated domains of BRCA1 and BRCA2. Prediction of the functional impact of all possible variants in BRCA1 and BRCA2 provides important information about the clinical relevance of variants in these genes.

KW - BRCA1 and BRCA2

KW - Functional evaluation

KW - In silico prediction

KW - Metapredictor

KW - VUS

UR - http://www.scopus.com/inward/record.url?scp=85048190898&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048190898&partnerID=8YFLogxK

U2 - 10.1038/s41436-018-0018-4

DO - 10.1038/s41436-018-0018-4

M3 - Article

SP - 1

EP - 10

JO - Genetics in Medicine

JF - Genetics in Medicine

SN - 1098-3600

ER -