{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Heiling HM"],"funding":["NIA NIH HHS","NCI NIH HHS","National Institutes of Health","NIH HHS"],"pagination":["ujae016"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC10946237"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["80(1)"],"pubmed_abstract":["Modern biomedical datasets are increasingly high-dimensional and exhibit complex correlation structures. Generalized linear mixed models (GLMMs) have long been employed to account for such dependencies. However, proper specification of the fixed and random effects in GLMMs is increasingly difficult in high dimensions, and computational complexity grows with increasing dimension of the random effects. We present a novel reformulation of the GLMM using a factor model decomposition of the random effects, enabling scalable computation of GLMMs in high dimensions by reducing the latent space from a large number of random effects to a smaller set of latent factors. We also extend our prior work to estimate model parameters using a modified Monte Carlo Expectation Conditional Minimization algorithm, allowing us to perform variable selection on both the fixed and random effects simultaneously. We show through simulation that through this factor model decomposition, our method can fit high-dimensional penalized GLMMs faster than comparable methods and more easily scale to larger dimensions not previously seen in existing approaches."],"journal":["Biometrics"],"pubmed_title":["Efficient computation of high-dimensional penalized generalized linear mixed models by latent factor modeling of the random effects."],"pmcid":["PMC10946237"],"funding_grant_id":["T32 CA106209","P50 CA257911","P50 CA058223","U01 CA274298","RO1 AG073259","R01 AG073259"],"pubmed_authors":["Li Q","Yeh JJ","Heiling HM","Ibrahim JG","Rashid NU","Peng XL"],"additional_accession":[]},"is_claimable":false,"name":"Efficient computation of high-dimensional penalized generalized linear mixed models by latent factor modeling of the random effects.","description":"Modern biomedical datasets are increasingly high-dimensional and exhibit complex correlation structures. Generalized linear mixed models (GLMMs) have long been employed to account for such dependencies. However, proper specification of the fixed and random effects in GLMMs is increasingly difficult in high dimensions, and computational complexity grows with increasing dimension of the random effects. We present a novel reformulation of the GLMM using a factor model decomposition of the random effects, enabling scalable computation of GLMMs in high dimensions by reducing the latent space from a large number of random effects to a smaller set of latent factors. We also extend our prior work to estimate model parameters using a modified Monte Carlo Expectation Conditional Minimization algorithm, allowing us to perform variable selection on both the fixed and random effects simultaneously. We show through simulation that through this factor model decomposition, our method can fit high-dimensional penalized GLMMs faster than comparable methods and more easily scale to larger dimensions not previously seen in existing approaches.","dates":{"release":"2024-01-01T00:00:00Z","publication":"2024 Jan","modification":"2025-04-04T01:29:12.315Z","creation":"2025-04-04T01:29:12.315Z"},"accession":"S-EPMC10946237","cross_references":{"pubmed":["38497825"],"doi":["10.1093/biomtc/ujae016"]}}