MOTIVATION: High-dimensional omics data are typically measured on limited sample sizes, which challenges model-based clustering methods such as Gaussian mixture models (GMMs), often leading to instability and poor generalization under complex mixture structures. To address these limitations, we developed Praxis-BGM, a natural-gradient variational inference framework for GMMs. Praxis-BGM enables semi-supervised transfer learning by incorporating an informative prior GMM estimated from large-scale reference data with robust cluster structures. The prior model can encode cluster-specific means, covariance structures, and structural connectivity patterns, and is updated using the target data with variational inference to improve clustering in small-sample settings. RESULTS: Using the Variational Online Newton (VON) algorithm, we derived natural-gradient updates for the standard parameters of GMMs. Implemented in the Python library JAX for accelerator-oriented computation, Praxis-BGM is computationally efficient and scalable. Across extensive simulations and two real-world applications-breast cancer bulk transcriptomics for subtype recovery and single-cell transcriptomics for cross-platform cell-type label transfer-Praxis-BGM improves posterior clustering performance, stability, and biological interpretability, even when priors are partially mismatched. AVAILABILITY AND IMPLEMENTATION: Praxis-BGM is freely available at https://github.com/ContiLab-usc/Praxis-BGM, and an archival version is available on Zenodo at https://doi.org/10.5281/zenodo.19657680.

MOTIVATION: Epidemiologic, clinical and translational studies are increasingly generating multiplatform omics data. Methods that can integrate across multiple high-dimensional data types while accounting for differential patterns are critical for uncovering novel associations and underlying relevant subgroups. RESULTS: We propose an integrative model to estimate latent unknown clusters (LUCID) aiming to both distinguish unique genomic, exposure and informative biomarkers/omic effects while jointly estimating subgroups relevant to the outcome of interest. Simulation studies indicate that we can obtain consistent estimates reflective of the true simulated values, accurately estimate subgroups and recapitulate subgroup-specific effects. We also demonstrate the use of the integrated model for future prediction of risk subgroups and phenotypes. We apply this approach to two real data applications to highlight the integration of genomic, exposure and metabolomic data. AVAILABILITY AND IMPLEMENTATION: The LUCID method is implemented through the LUCIDus R package available on CRAN (https://CRAN.R-project.org/package=LUCIDus). SUPPLEMENTARY INFORMATION: Supplementary materials are available at Bioinformatics online.

UNLABELLED: Identifying causal variants remains a key challenge in post-GWAS (genome-wide association study) era, as many GWAS single-nucleotide polymorphisms (SNPs) (including imputed ones) fall into non-coding regions, making it difficult to associate statistical significance with predicted functionality. Therefore, we created a web-based tool, Enlight, which overlays functional annotation information, such as histone modification states, methylation patterns, transcription factor binding sites, eQTL and higher-order chromosomal structure, to GWAS results. AVAILABILITY AND IMPLEMENTATION: Accessible by a Web browser at http://enlight.usc.edu.

BACKGROUND: Testing for marginal associations between numerous genetic variants and disease may miss complex relationships among variables (e.g., gene-gene interactions). Bayesian approaches can model multiple variables together and offer advantages over conventional model building strategies, including using existing biological evidence as modeling priors and acknowledging that many models may fit the data well. With many candidate variables, Bayesian approaches to variable selection rely on algorithms to approximate the posterior distribution of models, such as Markov-Chain Monte Carlo (MCMC). Unfortunately, MCMC is difficult to parallelize and requires many iterations to adequately sample the posterior. We introduce a scalable algorithm called PEAK that improves the efficiency of MCMC by dividing a large set of variables into related groups using a rooted graph that resembles a mountain peak. Our algorithm takes advantage of parallel computing and existing biological databases when available. RESULTS: By using graphs to manage a model space with more than 500,000 candidate variables, we were able to improve MCMC efficiency and uncover the true simulated causal variables, including a gene-gene interaction. We applied PEAK to a case-control study of childhood asthma with 2,521 genetic variants. We used an informative graph for oxidative stress derived from Gene Ontology and identified several variants in ERBB4, OXR1, and BCL2 with strong evidence for associations with childhood asthma. CONCLUSIONS: We introduced an extremely flexible analysis framework capable of efficiently performing Bayesian variable selection on many candidate variables. The PEAK algorithm can be provided with an informative graph, which can be advantageous when considering gene-gene interactions, or a symmetric graph, which simply divides the model space into manageable regions. The PEAK framework is compatible with various model forms, allowing for the algorithm to be configured for different study designs and applications, such as pathway or rare-variant analyses, by simple modifications to the model likelihood and proposal functions.

BACKGROUND: There has been considerable effort focused on developing efficient programs for tagging single-nucleotide polymorphisms (SNPs). Many of these programs do not account for potential reduced genomic coverage resulting from genotyping failures nor do they preferentially select SNPs based on functionality, which may be more likely to be biologically important. RESULTS: We have developed a user-friendly and efficient software program, Snagger, as an extension to the existing open-source software, Haploview, which uses pairwise r2 linkage disequilibrium between single nucleotide polymorphisms (SNPs) to select tagSNPs. Snagger distinguishes itself from existing SNP selection algorithms, including Tagger, by providing user options that allow for: (1) prioritization of tagSNPs based on certain characteristics, including platform-specific design scores, functionality (i.e., coding status), and chromosomal position, (2) efficient selection of SNPs across multiple populations, (3) selection of tagSNPs outside defined genomic regions to improve coverage and genotyping success, and (4) picking of surrogate tagSNPs that serve as backups for tagSNPs whose failure would result in a significant loss of data. Using HapMap genotype data from ten ENCODE regions and design scores for the Illumina platform, we show similar coverage and design score distribution and fewer total tagSNPs selected by Snagger compared to the web server Tagger. CONCLUSION: Snagger improves upon current available tagSNP software packages by providing a means for researchers to select tagSNPs that reliably capture genetic variation across multiple populations while accounting for significant genotyping failure risk and prioritizing on SNP-specific characteristics.