Winston Lian Chye Koh

The paper introduces MethylHDA, a method that combines helicase-dependent amplification with DNMT1 to preserve DNA methylation during targeted amplification.

This paper introduces a multimodal framework that links biosynthetic gene clusters and natural products by integrating genomic and chemical data through shared embeddings, enabling bidirectional retrieval and strain selection.

The paper integrates mass spectrometry and genomics data with language models to accelerate natural product discovery in microbial strains.

This paper compares two Megasphaera elsdenii strains, finding that JCM 35779 produces 5.8 times more hexanoic acid than JCM 1772 due to genomic differences affecting carbon flux.

The paper creates a fungal dataset of 518 strains from Singapore with taxonomic and visual phenotype data to aid in fungal classification and biodiversity studies.

This study uses protein language models to design sweeter and more thermostable brazzein variants, identifying mutations that improve heat resistance and sweetness.

This paper uses a mouse lemur transcriptomic atlas to identify new genes, splice junctions, and disease-related features, highlighting their relevance to primate biology and human disease.

Portal integrates large-scale single-cell datasets efficiently and accurately by preserving biological variation across samples, platforms, and species.

MapCell uses Siamese neural networks to learn a cell type distance metric, enabling annotation transfer across scRNA-seq datasets and identifying novel cell types.