I'm a computational biologist with a background in experimental physiology, combining biomedical research with method development for bioimage analysis, spatial omics, and multimodal representation learning.
I am especially interested in developing deep learning models to answer biological questions, supported by reproducible analysis and benchmarking infrastructure for rigorous evaluation and biological interpretation.
BenchRep — benchmarking representation learning for microscopy
An early-stage Python framework for reproducible benchmarking of microscopy representation-learning models, supporting both config-driven and programmatic workflows for training, prediction, evaluation, and provenance tracking.
VSTE — Variational Spatial Transcriptomics Encoder
A spatial-transcriptomics study and accompanying analysis repository evaluating VAE embeddings, downstream multimodal integration and clustering strategies, and classical CellProfiler image features for MERFISH-based cell-type deconvolution with DAPI and membrane imaging.
Associated publication: Evaluating integrative strategies for incorporating phenotypic features in spatial transcriptomics
sqY2H — screen quantification for composite plate images
An end-to-end image-analysis workflow I developed for registering composite yeast two-hybrid plate images, fitting well grids, segmenting and assigning colonies, and exporting continuous per-well growth measurements with quality-control and validation outputs.
Associated preprint: A scalable approach to resolving variants of uncertain significance
Offline, my non-scientific obsessions include chasing better hi-fi headphones and IEMs, biking in circles on gravel, hunting for movies I somehow missed, revisiting Star Trek, and negotiating with reality for more high-fantasy reading time — all while answering to my feline overlords: James Clerk Maxwell and ReLU 🐈.