Junjie Chen

This paper introduces ProtHMSO, a heuristic multi-site optimization framework using masked protein language models to design functional protein sequences more efficiently than existing methods.

PrefixProt uses prefix tuning to design proteins with controllable properties, outperforming other methods in efficiency and multi-objective control.

This paper introduces Enhancer-GAN, a GAN-based method for generating high-activity enhancer sequences, showing improved activity compared to benchmark datasets and identifying new motifs.

The paper presents a deep learning model, BERT-DPPIV, to identify DPP-IV inhibitory peptides and uncover design rules, achieving high accuracy and revealing restriction enzyme cutting sites.

This paper introduces iMFP-LG, a method combining protein language models and graph attention networks to identify multi-functional peptides, identifying 8 candidates and confirming one with antibacterial and anticancer effects.

STI is a transformer-based model for genome-scale genotype imputation that achieves high accuracy and handles multi-allelic variants effectively.

Mako detects complex structural variants using a graph-based pattern growth approach, achieving 70% validation rate on real data and identifying 15 CSV types including two novel ones.

The paper presents 64 haplotype-resolved human genomes using long-read sequencing and identifies structural variants, SV-eQTLs, and adaptive selection candidates.

This study evaluates differential privacy's effectiveness in defending against membership inference attacks on genomic machine learning models, showing a trade-off between privacy and accuracy.