Changiz Eslahchi

Evaluating microbial network inference methods: Moving beyond synthetic data with reproducibility-driven benchmarks

This paper evaluates six microbial network inference methods using real and generated data, showing that synthetic benchmarks can be misleading and proposing a reproducibility-focused framework for better evaluation.

Ancestra: A lineage-explicit simulator for benchmarking B-cell receptor repertoire and lineage inference methods

Ancestra simulates B-cell receptor repertoires with known lineage histories to benchmark clonal grouping and lineage reconstruction methods.

UnivAIRRse: A Unified Framework for Organizing and Comparing Adaptive Immune Receptor Repertoire Simulators

UnivAIRRse introduces a unified framework to organize and compare AIRR-seq simulators, identifying limitations and providing a tool for dynamic comparison.

ComplexMatrixComb: Predicting Drug Combination IC50 Doses via Complex Numbers and Matrix Factorization

ComplexMatrixComb predicts drug combination IC50 doses using complex numbers and matrix factorization, outperforming traditional methods and validating with experimental assays.

NN-RNALoc: neural network-based model for prediction of mRNA sub-cellular localization using distance-based sub-sequence profiles

NN-RNALoc predicts mRNA sub-cellular localization using distance-based sub-sequence profiles and protein-protein interaction data, achieving higher accuracy and faster computation than existing models.

PhiDsc: Protein functional mutation Identification by 3D Structure Comparison

PhiDsc identifies functionally relevant mutations in protein families by comparing 3D structures and mutation recurrence data, revealing known and novel hotspots in RAS and RHO families.

PSL-Recommender: Protein Subcellular Localization Prediction using Recommender System

PSL-Recommender improves protein subcellular localization prediction using a recommender system, outperforming existing methods on multiple datasets.