A statistical boosting framework for polygenic risk scores based on large-scale genotype data
Polygenic risk scores (PRS) evaluate the individual genetic liability to a certain trait and are expected to play an increasingly important role in clinical risk stratification. Most often, PRS are estimated based on summary statistics of univariate effects derived from genome-wide association studi...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-01-01
|
| Series: | Frontiers in Genetics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fgene.2022.1076440/full |
| _version_ | 1797956535785095168 |
|---|---|
| author | Hannah Klinkhammer Hannah Klinkhammer Christian Staerk Carlo Maj Carlo Maj Peter Michael Krawitz Andreas Mayr |
| author_facet | Hannah Klinkhammer Hannah Klinkhammer Christian Staerk Carlo Maj Carlo Maj Peter Michael Krawitz Andreas Mayr |
| author_sort | Hannah Klinkhammer |
| collection | DOAJ |
| description | Polygenic risk scores (PRS) evaluate the individual genetic liability to a certain trait and are expected to play an increasingly important role in clinical risk stratification. Most often, PRS are estimated based on summary statistics of univariate effects derived from genome-wide association studies. To improve the predictive performance of PRS, it is desirable to fit multivariable models directly on the genetic data. Due to the large and high-dimensional data, a direct application of existing methods is often not feasible and new efficient algorithms are required to overcome the computational burden regarding efficiency and memory demands. We develop an adapted component-wise L2-boosting algorithm to fit genotype data from large cohort studies to continuous outcomes using linear base-learners for the genetic variants. Similar to the snpnet approach implementing lasso regression, the proposed snpboost approach iteratively works on smaller batches of variants. By restricting the set of possible base-learners in each boosting step to variants most correlated with the residuals from previous iterations, the computational efficiency can be substantially increased without losing prediction accuracy. Furthermore, for large-scale data based on various traits from the UK Biobank we show that our method yields competitive prediction accuracy and computational efficiency compared to the snpnet approach and further commonly used methods. Due to the modular structure of boosting, our framework can be further extended to construct PRS for different outcome data and effect types—we illustrate this for the prediction of binary traits. |
| first_indexed | 2024-04-10T23:50:24Z |
| format | Article |
| id | doaj.art-9509bf080f0a42f3899f3de5a0a88e5d |
| institution | Directory Open Access Journal |
| issn | 1664-8021 |
| language | English |
| last_indexed | 2024-04-10T23:50:24Z |
| publishDate | 2023-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Genetics |
| spelling | doaj.art-9509bf080f0a42f3899f3de5a0a88e5d2023-01-10T18:53:59ZengFrontiers Media S.A.Frontiers in Genetics1664-80212023-01-011310.3389/fgene.2022.10764401076440A statistical boosting framework for polygenic risk scores based on large-scale genotype dataHannah Klinkhammer0Hannah Klinkhammer1Christian Staerk2Carlo Maj3Carlo Maj4Peter Michael Krawitz5Andreas Mayr6Institute for Medical Biometry, Informatics and Epidemiology, Medical Faculty, University of Bonn, Bonn, GermanyInstitute for Genomic Statistics and Bioinformatics, Medical Faculty, University of Bonn, Bonn, GermanyInstitute for Medical Biometry, Informatics and Epidemiology, Medical Faculty, University of Bonn, Bonn, GermanyInstitute for Genomic Statistics and Bioinformatics, Medical Faculty, University of Bonn, Bonn, GermanyCenter for Human Genetics, University of Marburg, Marburg, GermanyInstitute for Genomic Statistics and Bioinformatics, Medical Faculty, University of Bonn, Bonn, GermanyInstitute for Medical Biometry, Informatics and Epidemiology, Medical Faculty, University of Bonn, Bonn, GermanyPolygenic risk scores (PRS) evaluate the individual genetic liability to a certain trait and are expected to play an increasingly important role in clinical risk stratification. Most often, PRS are estimated based on summary statistics of univariate effects derived from genome-wide association studies. To improve the predictive performance of PRS, it is desirable to fit multivariable models directly on the genetic data. Due to the large and high-dimensional data, a direct application of existing methods is often not feasible and new efficient algorithms are required to overcome the computational burden regarding efficiency and memory demands. We develop an adapted component-wise L2-boosting algorithm to fit genotype data from large cohort studies to continuous outcomes using linear base-learners for the genetic variants. Similar to the snpnet approach implementing lasso regression, the proposed snpboost approach iteratively works on smaller batches of variants. By restricting the set of possible base-learners in each boosting step to variants most correlated with the residuals from previous iterations, the computational efficiency can be substantially increased without losing prediction accuracy. Furthermore, for large-scale data based on various traits from the UK Biobank we show that our method yields competitive prediction accuracy and computational efficiency compared to the snpnet approach and further commonly used methods. Due to the modular structure of boosting, our framework can be further extended to construct PRS for different outcome data and effect types—we illustrate this for the prediction of binary traits.https://www.frontiersin.org/articles/10.3389/fgene.2022.1076440/fullpolygenic risk score (PRS)high-dimensional datavariable selectionboostingGWAS—genome-wide association studyprediction |
| spellingShingle | Hannah Klinkhammer Hannah Klinkhammer Christian Staerk Carlo Maj Carlo Maj Peter Michael Krawitz Andreas Mayr A statistical boosting framework for polygenic risk scores based on large-scale genotype data Frontiers in Genetics polygenic risk score (PRS) high-dimensional data variable selection boosting GWAS—genome-wide association study prediction |
| title | A statistical boosting framework for polygenic risk scores based on large-scale genotype data |
| title_full | A statistical boosting framework for polygenic risk scores based on large-scale genotype data |
| title_fullStr | A statistical boosting framework for polygenic risk scores based on large-scale genotype data |
| title_full_unstemmed | A statistical boosting framework for polygenic risk scores based on large-scale genotype data |
| title_short | A statistical boosting framework for polygenic risk scores based on large-scale genotype data |
| title_sort | statistical boosting framework for polygenic risk scores based on large scale genotype data |
| topic | polygenic risk score (PRS) high-dimensional data variable selection boosting GWAS—genome-wide association study prediction |
| url | https://www.frontiersin.org/articles/10.3389/fgene.2022.1076440/full |
| work_keys_str_mv | AT hannahklinkhammer astatisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT hannahklinkhammer astatisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT christianstaerk astatisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT carlomaj astatisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT carlomaj astatisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT petermichaelkrawitz astatisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT andreasmayr astatisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT hannahklinkhammer statisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT hannahklinkhammer statisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT christianstaerk statisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT carlomaj statisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT carlomaj statisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT petermichaelkrawitz statisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata AT andreasmayr statisticalboostingframeworkforpolygenicriskscoresbasedonlargescalegenotypedata |