Md. Shamsuzzoha Bayzid

QRAFT efficiently infers species trees for up to a million taxa using coalescent-based methods, outperforming existing tools in speed and memory usage.

This paper introduces MirrorDNA, a DNA language model with a strand-equivariant tokenizer that improves prediction consistency across DNA strands.

QUANTA uses quantum annealing to improve species tree estimation from multi-locus data, achieving comparable accuracy to classical methods with significant speedup.

ASTRAL-X enables statistically consistent species tree inference for up to 300,000 taxa by improving scalability and reducing computational requirements.

This study compares the effects of incomplete lineage sorting and gene tree estimation error on species tree inference using simulated and real datasets.

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

STEQ is a fast and accurate distance-based method for estimating species trees from gene trees, maintaining competitive accuracy while being asymptotically faster than existing methods.

STELAR-X is a scalable triplet-based method for species tree inference that outperforms existing methods in speed and memory usage on large datasets.

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.

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.

This study evaluates how different gene tree rooting methods affect the accuracy of triplet-based species tree estimation compared to quartet-based methods like ASTRAL.

QT-WEAVER improves species tree accuracy by correcting quartet distribution errors from gene tree estimation, using a general-purpose method that adjusts quartet weights without explicit modeling of the system.

This study evaluates weighted quartet methods for species tree inference, finding that GTF-based approaches outperform ASTRAL and SVDquartets in accuracy.

This paper introduces wQFM-GDL, a method for species tree inference that models gene duplication and loss, outperforming existing methods on large datasets.

This paper introduces an alignment-free maximum likelihood method for phylogenetic tree construction and shows it is competitive with existing alignment-free tools.

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

This paper investigates species tree terraces caused by missing data, showing they can improve summary method accuracy and branch support computation.

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.

Pair-EGRET predicts protein-protein interaction sites using graph attention networks and protein language models, outperforming existing methods on experimental data.

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

This study conducts GWAS on 7 yield-related traits in 183 Bangladeshi rice landraces, identifying chromosomal regions and genes associated with grain shape and leaf traits.

This study introduces wQFM-DISCO, a method combining DISCO and wQFM to improve phylogenomic analyses by handling multicopy gene trees from gene duplication and loss events, showing higher accuracy than existing methods.

SAINT-Angle improves protein backbone torsion angle prediction using a self-attention network and transfer learning, outperforming existing methods on benchmark datasets.

QT-GILD improves species tree estimation by imputing missing quartets from incomplete gene trees using deep learning, enhancing accuracy despite data gaps.

The paper introduces an alignment-free method for phylogenetic distance estimation under a no-strand bias model, showing high accuracy in simulations and real data.

EGRET improves PPI site prediction using a graph attention network and transfer learning from pre-trained models.

The paper introduces wQFM, a method for estimating species trees from weighted quartets, showing it matches or improves upon existing methods in accuracy.

This paper introduces two machine learning methods for imputing missing values in distance matrices to construct phylogenetic trees, showing improved accuracy and scalability over existing techniques.

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