Mostofa Rafid Uddin

Localization of Macromolecules in Crowded Cellular Cryo-electron Tomograms from Extremely Sparse Labels

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

Deep Disentangled Representation Learning Reveals Neuron Subtype-Specific Nuclear Morphologies Across Aging in Mice and Humans

The paper develops a computational pipeline to link nuclear morphology to mRNA expression in neurons across aging in mice and humans, revealing subtype-specific morphology differences and age-related interactions.

Unsupervised Multi-scale Segmentation of Cellular Cryo-electron Tomograms with Stable Diffusion Foundation Model

The paper introduces an unsupervised method for segmenting subcellular structures in cryo-ET images using a Stable Diffusion model, achieving performance close to expert annotations.

DUAL: deep unsupervised simultaneous simulation and denoising for cryo-electron tomography

DUAL improves cryo-electron tomography by simultaneously denoising and simulating data without labeled training data, enhancing structural detection and annotation.

SAINT: Self-Attention AugmentedInception-Inside-Inception Network Improves ProteinSecondary Structure Prediction

SAINT improves Q8 protein secondary structure prediction using self-attention and Deep3I network, outperforming existing methods on benchmark datasets.

eMED-DNA: An in silico operating system for clinical medical data storage within the human genome

The paper introduces eMED-DNA, an in silico operating system for storing electronic health records within a sequenced genome to integrate genomic and clinical data in real-world practice.