Nico Pfeifer

The paper benchmarks generative models for synthetic bulk RNA-seq data, finding trade-offs between utility, biological preservation, and privacy risks across different methods.

This paper predicts drug response in epilepsy using multi-task gradient-boosted trees, identifying key genomic and clinical features that influence treatment outcomes.

This paper develops Bayesian generalized linear mixed models to estimate HLA adaptation in HIV-1 and finds that CXCR4-using variants are more adapted than CCR5-using variants.

The paper evaluates multi-view confounder detection methods for biomedical studies and finds that integrating multi-omics data improves confounding detection compared to single-modality approaches.

The paper develops a machine learning framework that uses phenotypic similarity to predict the functional effects of missense variants in voltage-gated sodium channels, achieving high accuracy and robustness.

This paper develops a multi-task learning model to predict the functional effects of voltage-gated potassium channel variants, achieving improved accuracy over standard methods.

The paper compares DNase I-Seq, ATAC-Seq, and NOMe-Seq to identify assay-specific features and their impact on chromatin accessibility and gene expression prediction.

This paper introduces CoMIK, a discriminative method for comparing variable-length genomic sequences, showing improved classification performance and interpretability.