Bangladesh University of Engineering and Technology

Single-cell perturbation prediction by conditional flow matching, using one cell-type-conditioned model in place of a separate model per cell type.

Antibody CDR design model fusing diffusion with a GFlowNet objective so binding energy is optimized during training rather than by post-hoc RL.

N-linked glycosylation site prediction stacking SVM, XGBoost, and KNN classifiers over ProtT5, ESM-2, and ProteinBERT sequence embeddings.

Protein-conditioned RNA sequence and structure co-design, refining flow-matched backbones against Lennard-Jones and folding free-energy terms.

Constrained RNA sequence design that turns an encoder-only RNA language model into a generator by greedy gradient search over nucleotides.

RNA language model that switches between nucleotide and byte-pair tokenization by input length, so one 117M encoder handles sequences of any length.