Yang Young Lu

PPI-Caliper controls false discovery rate in protein interaction prediction using conformal calibration without retraining models.

This study evaluates the reliability of protein sequence search using language models and finds that classical alignment methods are more robust than PLM-based approaches.

The paper improves the Tide search engine to handle large-scale mass spectrometry data efficiently, achieving faster speeds and lower memory usage than previous versions.

The paper introduces SONATA, a diagnostic method to detect artificial integrations in diagonal multimodal single-cell data integration, showing its effectiveness across simulated and real datasets.

The paper introduces Ledidi, a computational method for designing cis-regulatory elements by optimizing continuous sequences guided by deep learning models, validated experimentally for controlling transcriptional outcomes.

PhyloMix improves microbiome-trait association prediction by generating synthetic samples using phylogenetic relationships, outperforming existing augmentation methods.

The paper evaluates blastp's E-value and proposes an alternative significance analysis that provides more accurate alignment scores.

MIOSTONE models microbiome-trait associations using taxonomy-adaptive neural networks to predict and interpret microbial impacts on host traits.

BioMANIA provides a codeless, AI-driven bioinformatics pipeline that simplifies omics data analysis by interpreting natural language instructions and integrating existing tools.

The paper summarizes recent developments in the Crux toolkit for analyzing bottom-up proteomics data, including speedups, new features, and tools.

The paper introduces a proteomic technology called Miro to identify amino acid substitutions that affect protein function by using mistranslation and mass spectrometry.

DIAmeter is a library-free search engine for analyzing DIA mass spectrometry data using peptide sequences, supporting various isolation windows and post-translational modifications.

ACE integrates clustering and marker gene selection into a single deep learning workflow to identify discriminative and nonredundant gene panels.

SAINT is an alignment-free sequence comparison method using a Siamese triplet network for efficient and accurate taxonomy embedding and categorization.

CRAFT learns compact sequence representations for fast alignment-free comparison of genomic and metagenomic data, achieving efficient storage and query performance.

Ledidi designs genome edits to induce functional activity by optimizing predictions from a model, validated using Basenji for CTCF and JUND binding.

The paper trains machine learning models to predict gene expression in Plasmodium falciparum across three red blood cell stages using genomic and epigenomic data.

The paper introduces a regression model called Siren to improve peptide detection in DIA mass spectra by accounting for isotope peak interference.