Timothy Shin Heng Mak
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Timothy Shin Heng Mak has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Pak Chung Sham
Department of Psychiatry, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong SAR, China
7 shared publications - Shing Wan Choi
Department of Genetics and Genomic Sciences, Icahn School of Medicine
7 shared publications - Paul F. O’reilly
Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, USA
4 shared publications - Robert Milan Porsch
University of Hong Kong
4 shared publications - Sheilagh Hodgins
Département de psychiatrie, Université de Montréal, Montréal, Canada
4 shared publications A. Cree
Department of Forensic Mental Health Science, Institute of Psychiatry, King's College London, Box PO 23 De Crespigny Park, Denmark Hill, London SE5 8AF, UK
3 shared publications
Latest Publications
- Open Access
- 28 December 2022
- Published by Oxford University Press (OUP) in GigaScience
- Vol. 12
EraSOR: a software tool to eliminate inflation caused by sample overlap in polygenic score analyses
Background: Polygenic risk score (PRS) analyses are now routinely applied across biomedical research. However, as PRS studies grow in size, there is an increased risk of sample overlap between the genome-wide association study (GWAS) from which the PRS is derived and the “target sample,” in which PRSs are computed and hypotheses are tested. Despite the wide recognition of the sample overlap problem, its potential impact on the results from PRS studies has not yet been quantified, and no analytical solution has been provided. Findings: Here, we first conduct a comprehensive investigation into the scale of the sample overlap problem, finding that PRS results can be substantially inflated even in the presence of minimal overlap. Next, we introduce a method and software, EraSOR (Erase Sample Overlap and Relatedness), which eliminates the inflation caused by sample overlap (and close relatedness) in almost all settings tested here. Conclusions: EraSOR could be useful in PRS studies (with target sample >1,000) similar to those investigated here, either (i) to mitigate the potential effects of known or unknown intercohort overlap and close relatedness or (ii) as a sensitivity tool to highlight the possible presence of sample overlap before its direct removal, when possible, or else to provide a lower bound on PRS analysis results after accounting for potential sample overlap.
- 28 December 2022
- Published by Oxford University Press (OUP) in GigaScience
- Vol. 12
Open AccessEraSOR: a software tool to eliminate inflation caused by sample overlap in polygenic score analyses
Background: Polygenic risk score (PRS) analyses are now routinely applied across biomedical research. However, as PRS studies grow in size, there is an increased risk of sample overlap between the genome-wide association study (GWAS) from which the PRS is derived and the “target sample,” in which PRSs are computed and hypotheses are tested. Despite the wide recognition of the sample overlap problem, its potential impact on the results from PRS studies has not yet been quantified, and no analytical solution has been provided. Findings: Here, we first conduct a comprehensive investigation into the scale of the sample overlap problem, finding that PRS results can be substantially inflated even in the presence of minimal overlap. Next, we introduce a method and software, EraSOR (Erase Sample Overlap and Relatedness), which eliminates the inflation caused by sample overlap (and close relatedness) in almost all settings tested here. Conclusions: EraSOR could be useful in PRS studies (with target sample >1,000) similar to those investigated here, either (i) to mitigate the potential effects of known or unknown intercohort overlap and close relatedness or (ii) as a sensitivity tool to highlight the possible presence of sample overlap before its direct removal, when possible, or else to provide a lower bound on PRS analysis results after accounting for potential sample overlap.
- 25 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 10 (3) , 855-868
Two-Stage Auction Mechanism for Long-Term Participation in Crowdsourcing
Crowdsourcing has become an important tool to collect data for various artificial intelligence applications, and auction can be an effective way to allocate work and determine reward in a crowdsourcing platform. In this article, we focus on the crowdsourcing of small tasks such as image labeling and voice recording, where we face a number of challenges. First, workers have different limits on the amount of work they would be willing to do, and they may also misreport these limits in their bid for the work. Second, if the auction is repeated over time, unsuccessful workers may dropout of the system, reducing competition and diversity. To tackle these issues, we first extend the results of the celebrated Myerson’s optimal auction mechanism for a single-parameter bid to the case where the bid consists of the unit cost of work, the maximum amount of work one is willing to do, and the actual work completed. We show that a simple payment mechanism is sufficient to ensure a dominant strategy from the workers, and that this dominant strategy is robust to the true utility function of the workers. Second, we propose a novel, flexible work allocation mechanism, which allows the requester to balance between cost efficiency and equality. While cost minimization is obviously important, encouraging equality in the allocation of work increases the diversity of the workforce as well as promotes a long-term participation on the crowdsourcing platform. Our main results are proved analytically and validated through simulations.
- 25 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 10 (3) , 855-868
Two-Stage Auction Mechanism for Long-Term Participation in Crowdsourcing
Crowdsourcing has become an important tool to collect data for various artificial intelligence applications, and auction can be an effective way to allocate work and determine reward in a crowdsourcing platform. In this article, we focus on the crowdsourcing of small tasks such as image labeling and voice recording, where we face a number of challenges. First, workers have different limits on the amount of work they would be willing to do, and they may also misreport these limits in their bid for the work. Second, if the auction is repeated over time, unsuccessful workers may dropout of the system, reducing competition and diversity. To tackle these issues, we first extend the results of the celebrated Myerson’s optimal auction mechanism for a single-parameter bid to the case where the bid consists of the unit cost of work, the maximum amount of work one is willing to do, and the actual work completed. We show that a simple payment mechanism is sufficient to ensure a dominant strategy from the workers, and that this dominant strategy is robust to the true utility function of the workers. Second, we propose a novel, flexible work allocation mechanism, which allows the requester to balance between cost efficiency and equality. While cost minimization is obviously important, encouraging equality in the allocation of work increases the diversity of the workforce as well as promotes a long-term participation on the crowdsourcing platform. Our main results are proved analytically and validated through simulations.
- Open Access
EraSOR: Erase Sample Overlap in polygenic score analyses
Background: Polygenic risk score (PRS) analyses are now routinely applied in biomedical research, with great hope that they will aid in our understanding of disease aetiology and contribute to personalized medicine. The continued growth of multi-cohort genome-wide association studies (GWASs) and large-scale biobank projects has provided researchers with a wealth of GWAS summary statistics and individual-level data suitable for performing PRS analyses. However, as the size of these studies increase, the risk of inter-cohort sample overlap and close relatedness increases. Ideally sample overlap would be identified and removed directly, but this is typically not possible due to privacy laws or consent agreements. This sample overlap, whether known or not, is a major problem in PRS analyses because it can lead to inflation of type 1 error and, thus, erroneous conclusions in published work. Results: Here, for the first time, we report the scale of the sample overlap problem for PRS analyses by generating known sample overlap across sub-samples of the UK Biobank data, which we then use to produce GWAS and target data to mimic the effects of inter-cohort sample overlap. We demonstrate that inter-cohort overlap results in a significant and often substantial inflation in the observed PRS-trait association, coefficient of determination (R 2 ) and false-positive rate. This inflation can be high even when the absolute number of overlapping individuals is small if this makes up a notable fraction of the target sample. We develop and introduce EraSOR ( Era se S ample O verlap and R elatedness), a software for adjusting inflation in PRS prediction and association statistics in the presence of sample overlap or close relatedness between the GWAS and target samples. A key component of the EraSOR approach is inference of the degree of sample overlap from the intercept of a bivariate LD score regression applied to the GWAS and target data, making it powered in settings where both have sample sizes over 1,000 individuals. Through extensive benchmarking using UK Biobank and HapGen2 simulated genotype-phenotype data, we demonstrate that PRSs calculated using EraSOR-adjusted GWAS summary statistics are robust to inter-cohort overlap in a wide range of realistic scenarios and are even robust to high levels of residual genetic and environmental stratification. Conclusion: The results of all PRS analyses for which sample overlap cannot be definitively ruled out should be considered with caution given high type 1 error observed in the presence of even low overlap between base and target cohorts. Given the strong performance of EraSOR in eliminating inflation caused by sample overlap in PRS studies with large (>5k) target samples, we recommend that EraSOR be used in all future such PRS studies to mitigate the potential effects of inter-cohort overlap and close relatedness.
Open AccessEraSOR: Erase Sample Overlap in polygenic score analyses
Background: Polygenic risk score (PRS) analyses are now routinely applied in biomedical research, with great hope that they will aid in our understanding of disease aetiology and contribute to personalized medicine. The continued growth of multi-cohort genome-wide association studies (GWASs) and large-scale biobank projects has provided researchers with a wealth of GWAS summary statistics and individual-level data suitable for performing PRS analyses. However, as the size of these studies increase, the risk of inter-cohort sample overlap and close relatedness increases. Ideally sample overlap would be identified and removed directly, but this is typically not possible due to privacy laws or consent agreements. This sample overlap, whether known or not, is a major problem in PRS analyses because it can lead to inflation of type 1 error and, thus, erroneous conclusions in published work. Results: Here, for the first time, we report the scale of the sample overlap problem for PRS analyses by generating known sample overlap across sub-samples of the UK Biobank data, which we then use to produce GWAS and target data to mimic the effects of inter-cohort sample overlap. We demonstrate that inter-cohort overlap results in a significant and often substantial inflation in the observed PRS-trait association, coefficient of determination (R 2 ) and false-positive rate. This inflation can be high even when the absolute number of overlapping individuals is small if this makes up a notable fraction of the target sample. We develop and introduce EraSOR ( Era se S ample O verlap and R elatedness), a software for adjusting inflation in PRS prediction and association statistics in the presence of sample overlap or close relatedness between the GWAS and target samples. A key component of the EraSOR approach is inference of the degree of sample overlap from the intercept of a bivariate LD score regression applied to the GWAS and target data, making it powered in settings where both have sample sizes over 1,000 individuals. Through extensive benchmarking using UK Biobank and HapGen2 simulated genotype-phenotype data, we demonstrate that PRSs calculated using EraSOR-adjusted GWAS summary statistics are robust to inter-cohort overlap in a wide range of realistic scenarios and are even robust to high levels of residual genetic and environmental stratification. Conclusion: The results of all PRS analyses for which sample overlap cannot be definitively ruled out should be considered with caution given high type 1 error observed in the presence of even low overlap between base and target cohorts. Given the strong performance of EraSOR in eliminating inflation caused by sample overlap in PRS studies with large (>5k) target samples, we recommend that EraSOR be used in all future such PRS studies to mitigate the potential effects of inter-cohort overlap and close relatedness.
- Top Cited Papers
- 24 July 2020
- Published by Springer Nature in Nature Protocols
- Vol. 15 (9) , 2759-2772
Tutorial: a guide to performing polygenic risk score analyses
A polygenic score (PGS) or polygenic risk score (PRS) is an estimate of an individual’s genetic liability to a trait or disease, calculated according to their genotype profile and relevant genome-wide association study (GWAS) data. While present PRSs typically explain only a small fraction of trait variance, their correlation with the single largest contributor to phenotypic variation—genetic liability—has led to the routine application of PRSs across biomedical research. Among a range of applications, PRSs are exploited to assess shared etiology between phenotypes, to evaluate the clinical utility of genetic data for complex disease and as part of experimental studies in which, for example, experiments are performed that compare outcomes (e.g., gene expression and cellular response to treatment) between individuals with low and high PRS values. As GWAS sample sizes increase and PRSs become more powerful, PRSs are set to play a key role in research and stratified medicine. However, despite the importance and growing application of PRSs, there are limited guidelines for performing PRS analyses, which can lead to inconsistency between studies and misinterpretation of results. Here, we provide detailed guidelines for performing and interpreting PRS analyses. We outline standard quality control steps, discuss different methods for the calculation of PRSs, provide an introductory online tutorial, highlight common misconceptions relating to PRS results, offer recommendations for best practice and discuss future challenges.
- 24 July 2020
- Published by Springer Nature in Nature Protocols
- Vol. 15 (9) , 2759-2772
Top Cited PapersTutorial: a guide to performing polygenic risk score analyses
A polygenic score (PGS) or polygenic risk score (PRS) is an estimate of an individual’s genetic liability to a trait or disease, calculated according to their genotype profile and relevant genome-wide association study (GWAS) data. While present PRSs typically explain only a small fraction of trait variance, their correlation with the single largest contributor to phenotypic variation—genetic liability—has led to the routine application of PRSs across biomedical research. Among a range of applications, PRSs are exploited to assess shared etiology between phenotypes, to evaluate the clinical utility of genetic data for complex disease and as part of experimental studies in which, for example, experiments are performed that compare outcomes (e.g., gene expression and cellular response to treatment) between individuals with low and high PRS values. As GWAS sample sizes increase and PRSs become more powerful, PRSs are set to play a key role in research and stratified medicine. However, despite the importance and growing application of PRSs, there are limited guidelines for performing PRS analyses, which can lead to inconsistency between studies and misinterpretation of results. Here, we provide detailed guidelines for performing and interpreting PRS analyses. We outline standard quality control steps, discuss different methods for the calculation of PRSs, provide an introductory online tutorial, highlight common misconceptions relating to PRS results, offer recommendations for best practice and discuss future challenges.
- Open Access
A guide to performing Polygenic Risk Score analyses
The application of polygenic risk scores (PRS) has become routine across genetic research. Among a range of applications, PRS are exploited to assess shared aetiology between phenotypes, to evaluate the predictive power of genetic data for use in clinical settings, and as part of experimental studies in which, for example, experiments are performed on individuals, or their biological samples (eg. tissues, cells), at the tails of the PRS distribution and contrasted. As GWAS sample sizes increase and PRS become more powerful, they are set to play a key role in personalised medicine. However, despite the growing application and importance of PRS, there are limited guidelines for performing PRS analyses, which can lead to inconsistency between studies and misinterpretation of results. Here we provide detailed guidelines for performing polygenic risk score analyses relevant to different methods for their calculation, outlining standard quality control steps and offering recommendations for best-practice. We also discuss different methods for the calculation of PRS, common misconceptions regarding the interpretation of results and future challenges.
Open AccessA guide to performing Polygenic Risk Score analyses
The application of polygenic risk scores (PRS) has become routine across genetic research. Among a range of applications, PRS are exploited to assess shared aetiology between phenotypes, to evaluate the predictive power of genetic data for use in clinical settings, and as part of experimental studies in which, for example, experiments are performed on individuals, or their biological samples (eg. tissues, cells), at the tails of the PRS distribution and contrasted. As GWAS sample sizes increase and PRS become more powerful, they are set to play a key role in personalised medicine. However, despite the growing application and importance of PRS, there are limited guidelines for performing PRS analyses, which can lead to inconsistency between studies and misinterpretation of results. Here we provide detailed guidelines for performing polygenic risk score analyses relevant to different methods for their calculation, outlining standard quality control steps and offering recommendations for best-practice. We also discuss different methods for the calculation of PRS, common misconceptions regarding the interpretation of results and future challenges.
- Open Access
- 18 July 2018
- Published by Springer Nature in Scientific Reports
- Vol. 8 (1) , 10846
Coverage and diagnostic yield of Whole Exome Sequencing for the Evaluation of Cases with Dilated and Hypertrophic Cardiomyopathy
Targeted next generation sequencing of gene panels has become a popular tool for the genetic diagnosis of hypertrophic (HCM) and dilated cardiomyopathy (DCM). However, it is uncertain whether the use of Whole Exome Sequencing (WES) represents a more effective approach for diagnosis of cases with HCM and DCM. In this study, we performed indirect comparisons of the coverage and diagnostic yield of WES on genes and variants related to HCM and DCM versus 4 different commercial gene panels using 40 HCM and DCM patients, assuming perfect coverage in those panels. We identified 6 pathogenic or likely pathogenic among 14 HCM patients (diagnostic yield 43%). 3 pathogenic or likely pathogenic were found among the 26 DCM patients (diagnostic yield 12%). The coverage was similar to that of four existing commercial gene panels due to the clustering of mutation within MYH7, MYBPC3, TPM1, TNT2, and TTN. Moreover, the coverage of WES appeared inadequate for TNNI3 and PLN. We conclude that most of the pathogenic variants for HCM and DCM can be found within a small number of genes which were covered by all the commercial gene panels, and the application of WES did not increase diagnostic yield.
- 18 July 2018
- Published by Springer Nature in Scientific Reports
- Vol. 8 (1) , 10846
Open AccessCoverage and diagnostic yield of Whole Exome Sequencing for the Evaluation of Cases with Dilated and Hypertrophic Cardiomyopathy
Targeted next generation sequencing of gene panels has become a popular tool for the genetic diagnosis of hypertrophic (HCM) and dilated cardiomyopathy (DCM). However, it is uncertain whether the use of Whole Exome Sequencing (WES) represents a more effective approach for diagnosis of cases with HCM and DCM. In this study, we performed indirect comparisons of the coverage and diagnostic yield of WES on genes and variants related to HCM and DCM versus 4 different commercial gene panels using 40 HCM and DCM patients, assuming perfect coverage in those panels. We identified 6 pathogenic or likely pathogenic among 14 HCM patients (diagnostic yield 43%). 3 pathogenic or likely pathogenic were found among the 26 DCM patients (diagnostic yield 12%). The coverage was similar to that of four existing commercial gene panels due to the clustering of mutation within MYH7, MYBPC3, TPM1, TNT2, and TTN. Moreover, the coverage of WES appeared inadequate for TNNI3 and PLN. We conclude that most of the pathogenic variants for HCM and DCM can be found within a small number of genes which were covered by all the commercial gene panels, and the application of WES did not increase diagnostic yield.
- Open Access
KS-Burden: Assessing distributional differences of rare variants in dichotomous traits
A number of rare variant tests have been developed to explore the effect of low frequency genetic variations on complex phenotypes. However, an often neglected aspect in these tests is the position of genetic variations. Here we are proposing a way to assess the differences in spatial organization of rare variants by assessing their distributional differences between affected and unaffected subjects. To do so, we have formulated an adaptation of the well know Kolmogorov-Smirnov (KS) test, combining both KS and a simple gene burden approach, called KS-Burden. The performance of our test was evaluated under a comprehensive simulations framework using real data and various scenarios. Our results show that the KS-Burden test is able to outperform the commonly used SKAT-O test, as well as others, in the presents of clusters of causal variants within a genomic region. Furthermore, our test is able to maintain competitive statistical power in scenarios unfavorable to its original assumptions. Hence, the KS-Burden test is a valuable alternative to existing tests and provides better statistical power in the presents of causal clusters within a gene.
Open AccessKS-Burden: Assessing distributional differences of rare variants in dichotomous traits
A number of rare variant tests have been developed to explore the effect of low frequency genetic variations on complex phenotypes. However, an often neglected aspect in these tests is the position of genetic variations. Here we are proposing a way to assess the differences in spatial organization of rare variants by assessing their distributional differences between affected and unaffected subjects. To do so, we have formulated an adaptation of the well know Kolmogorov-Smirnov (KS) test, combining both KS and a simple gene burden approach, called KS-Burden. The performance of our test was evaluated under a comprehensive simulations framework using real data and various scenarios. Our results show that the KS-Burden test is able to outperform the commonly used SKAT-O test, as well as others, in the presents of clusters of causal variants within a genomic region. Furthermore, our test is able to maintain competitive statistical power in scenarios unfavorable to its original assumptions. Hence, the KS-Burden test is a valuable alternative to existing tests and provides better statistical power in the presents of causal clusters within a gene.
- Open Access
Trans-ethnic polygenic analysis supports genetic overlaps of lumbar disc degeneration with height, body mass index, and bone mineral density
Lumbar disc degeneration (LDD) is age-related break-down in the fibrocartilaginous joints between lumbar vertebrae. It is a major cause of low back pain and is conventionally assessed by magnetic resonance imaging (MRI). Like most other complex traits, LDD is likely polygenic and influenced by both genetic and environmental factors. However, genome-wide association studies (GWASs) of LDD have uncovered few susceptibility loci due to the limited sample size. Previous epidemiology studies of LDD also reported multiple heritable risk factors, including height, body mass index (BMI), bone mineral density (BMD), lipid levels, etc. Genetics can help elucidate causality between traits and suggest loci with pleiotropic effects. One such approach is polygenic score (PGS) which summarizes the effect of multiple variants by the summation of alleles weighted by estimated effects from GWAS. To investigate genetic overlaps of LDD and related heritable risk factors, we calculated the PGS of height, BMI, BMD and lipid levels in a Chinese population-based cohort with spine MRI examination and a Japanese case-control cohort of lumbar disc herniation (LDH) requiring surgery. Because most large-scale GWASs were done in European populations, PGS of corresponding traits were created using weights from European GWASs. We calibrated their prediction performance in independent Chinese samples, then tested associations with MRI-derived LDD scores and LDH affection status. The PGS of height, BMI, BMD and lipid levels were strongly associated with respective phenotypes in Chinese, although phenotype variances explained were lower than in Europeans which would reduce the power to detect genetic overlaps. Despite of this, the PGS of BMI and lumbar spine BMD were significantly associated with LDD scores; and the PGS of height was associated with the increased the liability of LDH. Furthermore, linkage disequilibrium score regression suggested that, osteoarthritis, another degenerative disorder that shares common features with LDD, also showed genetic correlations with height, BMI and BMD. The findings suggest a common key contribution of biomechanical stress to the pathogenesis of LDD and will direct the future search for pleiotropic genes.
Open AccessTrans-ethnic polygenic analysis supports genetic overlaps of lumbar disc degeneration with height, body mass index, and bone mineral density
Lumbar disc degeneration (LDD) is age-related break-down in the fibrocartilaginous joints between lumbar vertebrae. It is a major cause of low back pain and is conventionally assessed by magnetic resonance imaging (MRI). Like most other complex traits, LDD is likely polygenic and influenced by both genetic and environmental factors. However, genome-wide association studies (GWASs) of LDD have uncovered few susceptibility loci due to the limited sample size. Previous epidemiology studies of LDD also reported multiple heritable risk factors, including height, body mass index (BMI), bone mineral density (BMD), lipid levels, etc. Genetics can help elucidate causality between traits and suggest loci with pleiotropic effects. One such approach is polygenic score (PGS) which summarizes the effect of multiple variants by the summation of alleles weighted by estimated effects from GWAS. To investigate genetic overlaps of LDD and related heritable risk factors, we calculated the PGS of height, BMI, BMD and lipid levels in a Chinese population-based cohort with spine MRI examination and a Japanese case-control cohort of lumbar disc herniation (LDH) requiring surgery. Because most large-scale GWASs were done in European populations, PGS of corresponding traits were created using weights from European GWASs. We calibrated their prediction performance in independent Chinese samples, then tested associations with MRI-derived LDD scores and LDH affection status. The PGS of height, BMI, BMD and lipid levels were strongly associated with respective phenotypes in Chinese, although phenotype variances explained were lower than in Europeans which would reduce the power to detect genetic overlaps. Despite of this, the PGS of BMI and lumbar spine BMD were significantly associated with LDD scores; and the PGS of height was associated with the increased the liability of LDH. Furthermore, linkage disequilibrium score regression suggested that, osteoarthritis, another degenerative disorder that shares common features with LDD, also showed genetic correlations with height, BMI and BMD. The findings suggest a common key contribution of biomechanical stress to the pathogenesis of LDD and will direct the future search for pleiotropic genes.
- Open Access
Polygenic scores for UK Biobank scale data
Polygenic scores (PGS) are estimated scores representing the genetic tendency of an individual for a disease or trait and have become an indispensible tool in a variety of analyses. Typically they are linear combination of the genotypes of a large number of SNPs, with the weights calculated from an external source, such as summary statistics from large meta-analyses. Recently cohorts with genetic data have become very large, such that it would be a waste if the raw data were not made use of in constructing PGS. Making use of raw data in calculating PGS, however, presents us with problems of overfitting. Here we discuss the essence of overfitting as applied in PGS calculations and highlight the difference between overfitting due to the overlap between the target and the discovery data (OTD), and overfitting due to the overlap between the target the the validation data (OTV). We propose two methods — cross prediction and split validation — to overcome OTD and OTV respectively. Using these two methods, PGS can be calculated using raw data without overfitting. We show that PGSs thus calculated have better predictive power than those using summary statistics alone for six phenotypes in the UK Biobank data.
Open AccessPolygenic scores for UK Biobank scale data
Polygenic scores (PGS) are estimated scores representing the genetic tendency of an individual for a disease or trait and have become an indispensible tool in a variety of analyses. Typically they are linear combination of the genotypes of a large number of SNPs, with the weights calculated from an external source, such as summary statistics from large meta-analyses. Recently cohorts with genetic data have become very large, such that it would be a waste if the raw data were not made use of in constructing PGS. Making use of raw data in calculating PGS, however, presents us with problems of overfitting. Here we discuss the essence of overfitting as applied in PGS calculations and highlight the difference between overfitting due to the overlap between the target and the discovery data (OTD), and overfitting due to the overlap between the target the the validation data (OTV). We propose two methods — cross prediction and split validation — to overcome OTD and OTV respectively. Using these two methods, PGS can be calculated using raw data without overfitting. We show that PGSs thus calculated have better predictive power than those using summary statistics alone for six phenotypes in the UK Biobank data.
- Top Cited PapersOpen Access
- 8 May 2017
- Published by Wiley in Genetic Epidemiology
- Vol. 41 (6) , 469-480
Polygenic scores via penalized regression on summary statistics
Polygenic scores (PGS) summarize the genetic contribution of a person's genotype to a disease or phenotype. They can be used to group participants into different risk categories for diseases, and are also used as covariates in epidemiological analyses. A number of possible ways of calculating PGS have been proposed, and recently there is much interest in methods that incorporate information available in published summary statistics. As there is no inherent information on linkage disequilibrium (LD) in summary statistics, a pertinent question is how we can use LD information available elsewhere to supplement such analyses. To answer this question, we propose a method for constructing PGS using summary statistics and a reference panel in a penalized regression framework, which we call lassosum. We also propose a general method for choosing the value of the tuning parameter in the absence of validation data. In our simulations, we showed that pseudovalidation often resulted in prediction accuracy that is comparable to using a dataset with validation phenotype and was clearly superior to the conservative option of setting the tuning parameter of lassosum to its lowest value. We also showed that lassosum achieved better prediction accuracy than simple clumping and P‐value thresholding in almost all scenarios. It was also substantially faster and more accurate than the recently proposed LDpred.
- 8 May 2017
- Published by Wiley in Genetic Epidemiology
- Vol. 41 (6) , 469-480
Top Cited PapersOpen AccessPolygenic scores via penalized regression on summary statistics
Polygenic scores (PGS) summarize the genetic contribution of a person's genotype to a disease or phenotype. They can be used to group participants into different risk categories for diseases, and are also used as covariates in epidemiological analyses. A number of possible ways of calculating PGS have been proposed, and recently there is much interest in methods that incorporate information available in published summary statistics. As there is no inherent information on linkage disequilibrium (LD) in summary statistics, a pertinent question is how we can use LD information available elsewhere to supplement such analyses. To answer this question, we propose a method for constructing PGS using summary statistics and a reference panel in a penalized regression framework, which we call lassosum. We also propose a general method for choosing the value of the tuning parameter in the absence of validation data. In our simulations, we showed that pseudovalidation often resulted in prediction accuracy that is comparable to using a dataset with validation phenotype and was clearly superior to the conservative option of setting the tuning parameter of lassosum to its lowest value. We also showed that lassosum achieved better prediction accuracy than simple clumping and P‐value thresholding in almost all scenarios. It was also substantially faster and more accurate than the recently proposed LDpred.