Abrar Rahman Abir

ST-DISTAL improves cell-type deconvolution in spatial transcriptomics by integrating spatial and molecular information through a dual-branch graph convolutional framework.

ConnectoFM is a foundation model for connectomics that outperforms existing methods in low-data regimes and improves accuracy for critical biological structures.

MExConn is a multi-expert U-Net framework for segmenting multiple organelles in EM images, outperforming baselines and offering mechanistic interpretability.

MACS improves connectomics segmentation by adapting across multiple domains and using active learning, achieving significant performance gains over existing methods.

UnCOT-AD improves Alzheimer's Disease prediction by integrating unpaired multi-omics data using a novel deep learning framework with cycle consistency and contrastive learning.

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.

RNA-EFM uses energy-based flow matching to design RNA sequences and structures conditioned on protein targets, achieving better performance than existing methods.

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.

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