Mahdi Pourmirzaei

This paper introduces GCP-VQVAE, a geometry-complete tokenizer for protein structures, achieving state-of-the-art performance on benchmark datasets and demonstrating robust generalization to unseen proteins.

This paper benchmarks sequence-based models for predicting protein-ligand binding sites across different ligand availability regimes, showing improved performance with ligand conditioning and chemical language model integration.

The paper introduces two sequence-based models to predict kinase-specific phosphorylation sites, showing improved performance and zero-shot capability.

This paper introduces a unified model combining a protein language model with an autoregressive transformer to predict protein-ligand binding sites, achieving broad ligand coverage and improved performance for underrepresented ligands.

This paper extends Prot2Token to align protein language models for diverse prediction tasks, improving performance across multiple applications through pre-training and integration with ESM-2.

This paper introduces S-PLM, a 3D structure-aware protein language model that improves prediction tasks by integrating structural information through efficient fine-tuning.

S-PLM improves protein language models by integrating 3D structure information through contrastive learning, outperforming sequence-only models in clustering and classification tasks.

Prot2Token is a multi-task framework that uses autoregressive language modeling for protein prediction tasks, achieving performance comparable to specialized models and enabling integration with large language models.