Abrar Rahman Abir
I am Abrar Rahman Abir, a CS PhD student at University of California, Berkeley,advised by Dr. Allon Wagner. From the very beginning of my undergraduate journey, I have been deeply passionate about Computational Biology and Machine Learning. My profound fascination for Mathematics drives me to explore the vast realm of Machine Learning, and I am particularly excited by the challenge of designing novel machine learning methods to address complex biological problems.
My research spans multiple domains, with a strong publication record :
- Machine Learning
- Generative Models for Biology
- Single-cell and Spatial transcriptomics
- Single-cell and Spatial metabolomics
Publications
- AbFlowNet: Optimizing Antibody-Antigen Binding Energy via Diffusion-GFlowNet Fusion
- UnCOT-AD: Unpaired Cross-Omics Translation Enables Multi-Omics Integration for Alzheimer’s Disease Prediction
- LOCAS: Multi-label mRNA Localization with Supervised Contrastive Learning
- GTG-ACO: Graph Transformer Guided Ant Colony Optimization for Learning Heuristics and Pheromone Dynamics for Combinatorial Optimization
- DeepRNA-Twist: Language Model Guided RNA Torsion Angle Prediction with Attention-Inception Network
- BioLLMNet: Enhancing RNA-Interaction Prediction with a Specialized Cross-LLM Transformation Network
- RNA-EFM: Energy-based Flow Matching for Protein-conditioned RNA Sequence-Structure Co-design
- DCGAT-DTI: Dynamic Cross Graph Attention Network for Drug-Target Interaction Prediction
- CFM-GP: Unified Conditional Flow Matching to Learn Gene Perturbation Across Cell Types
- Use of CPT and Other Parameters for Estimating Soil Unit Weight using Optimised Machine Learning Models
- A Machine Learning Based Novel Approach of Predicting International Roughness Index (IRI) from Traffic Characteristics using Random Forest Regression
- ST-DISTAL: Learning Spatial Transcriptomics Deconvolution via Spectral-Spatial Dual-Branch GCNs with Distributional Alignment
- MACS: Multi-Domain Adaptation Facilitates Accurate Connectomics Segmentation
- MExConn: A Mechanistically Interpretable Multi-Expert Framework for Multi-Organelle Segmentation in Connectomics
- MOXGATE: Modality-Aware Cross-Attention for Multi-Omic Gastrointestinal Cancer Sub-type Classification
Multi-Label Contrastive Learning Augmented mRNA Subcellular Localization Prediction Leveraging RNA Language Model
Education
University of California, Berkeley
PhD in Computer Science (Fall 2026 - Present)