Chayaporn Suphavilai

Addressing pandemic-wide systematic errors in the SARS-CoV-2 phylogeny

The paper reprocesses SARS-CoV-2 genomic data to correct systematic errors in phylogenies using a novel assembly tool, resulting in a cleaner global tree of 4.47 million samples.

Direct high-throughput deconvolution of non-canonical bases via nanopore sequencing and bootstrapped learning

This paper demonstrates direct high-throughput sequencing of non-canonical bases using nanopore sequencing and bootstrapped learning, achieving high accuracy and specificity in basecalling.

MetageNN: a memory-efficient neural network taxonomic classifier robust to sequencing errors and missing genomes

MetageNN improves taxonomic classification of long-read sequencing data by using a neural network that is robust to sequencing errors and missing genomes, outperforming existing methods in sensitivity and speed.

Network-guided supervised learning on gene expression using a graph convolutional neural network

This paper introduces Grandline, a graph convolutional neural network that integrates gene expression and protein interaction networks to predict phenotypes with high accuracy and interpretability.

Predicting heterogeneity in clone-specific therapeutic vulnerabilities using single-cell transcriptomic signatures

The paper uses single-cell RNA-seq data to predict drug responses and combinations by analyzing intra-tumor transcriptomic heterogeneity in cancer cell lines.

TUGDA: Task uncertainty guided domain adaptation for robust generalization of cancer drug response prediction from in vitro to in vivo settings

TUGDA reduces negative transfer in drug response prediction by adapting in vitro models to in vivo settings through uncertainty-guided domain adaptation.

Predicting Cancer Drug Response Using a Recommender System

This paper introduces a recommender system approach to predict cancer drug responses using pharmacogenomic data and demonstrates its effectiveness across multiple datasets.

ConsensusDriver Improves Upon Individual Algorithms For Predicting Driver Alterations In Different Cancer Types And Individual Patients -- A Toolbox For Precision Oncology

ConsensusDriver improves driver alteration prediction by combining multiple methods, increasing sensitivity by 20% across cancer types and individual patients.