Md Toki Tahmid

The paper uses sparse autoencoders to analyze BiRNA-BERT, revealing biologically meaningful features and improving RNA-type classification.

BiRNA-BERT is a Transformer-based model pretrained on RNA sequences using adaptive tokenization, achieving state-of-the-art results in long-sequence tasks and efficient performance on short sequences.

TomoPicker is a macromolecule localization method for cryo-ET tomograms that achieves comparable performance to supervised methods with significantly fewer annotations.

DeepRNA-Twist predicts RNA torsion and pseudo-torsion angles from sequence using a deep learning framework with language model embeddings and an attention-inception network, achieving state-of-the-art accuracy.

This paper introduces wQFM-TREE, a method for species tree inference that improves accuracy and scalability over existing methods like ASTRAL and wQFM.

GraFusionNet improves graph classification by integrating node, edge, and semantic features, achieving better performance on toxicity prediction tasks.

BioLLMNet improves RNA-interaction prediction by combining multi-modal language model features using a gating mechanism, achieving state-of-the-art results in RNA-protein, RNA-small molecule, and RNA-RNA interactions.

This paper presents a compartmental-based simulation model for noninvasive mechanical ventilation to improve ventilator design and respiratory care in resource-limited settings.

EmbedSimScore improves protein similarity analysis by integrating structural and contextual embeddings to capture both local and global features.

LOCAS improves mRNA subcellular localization by using supervised contrastive learning and a multi-label classification head, achieving state-of-the-art performance.

RNA-DCGen generates RNA sequences with specific structural or functional properties using a language model, outperforming random baselines in structure prediction.