Hangzhou Institute of Medicine, CAS

Codon language model for mRNA prediction and coding sequence design, steering synonymous codon choice with a swappable host usage prior at inference.

Multimodal single-cell foundation model whose multiway Transformer jointly models scRNA-seq and scATAC-seq from RNA-only, ATAC-only, or paired inputs.

Structure-conditioned fine-tune of ESM2 for protein mutation-effect prediction, matching ESM3-level accuracy after roughly an hour of fine-tuning.

De novo antibiotic design framework coupling a 6.4B-parameter protein language model with reinforcement learning to generate antimicrobial peptides.

Single-cell ATAC-seq foundation model pretrained on 2.8 million cells across 1.15 million chromatin regions via masked peak reconstruction.

Nanopore basecaller built on a Squeezeformer encoder, turning raw ion-current signal into DNA at 93.97% average read identity across 11 datasets.

Discrete diffusion model that generates RNA secondary structure contact maps as pixel-wise segmentation, conditioned on RNA-FM and UFold features.

Protein language model ranking stabilizing mutations with no assay data, trained jointly to predict the growth temperature of each sequence's host.