Pranam Chatterjee

This paper introduces PepRePs, a system using peptide-retargeted phosphatases to selectively manipulate protein phosphorylation, showing reduced phosphorylation and downstream effects on cellular processes.

This paper develops programmable protein degraders (uAbs) that selectively knock down pathogenic β-catenin in vitro and in vivo, reducing tumor cell survival and proliferation.

This paper presents a high-throughput platform for screening peptide-guided protein degraders in mammalian cells, demonstrating their ability to reduce target protein abundance and function.

PeptiVerse provides a unified platform for predicting therapeutic peptide properties using both amino acid sequences and SMILES-based representations, supporting early-stage development and generative design.

AptaBLE is a deep learning framework that predicts aptamer-protein binding and generates novel aptamers with high specificity and low Kd values.

This paper develops a foundation model for microbial growth dynamics using self-supervised learning on diverse datasets, enabling accurate reconstruction and prediction of growth patterns and antibiotic responses.

The paper introduces SaprotHub, a platform that simplifies training and sharing of protein language models, enabling non-experts to build and collaborate on customized models.

This study introduces AMP-Diffusion, a generative model for designing antimicrobial peptides, which produces candidates with broad-spectrum antibacterial activity and low toxicity.

Metalorian generates heavy metal-binding peptides using a diffusion model guided by a classifier, producing peptides with controlled specificity and reduced length compared to known proteins.

DPAC predicts and designs protein-DNA interactions using sequence-based contrastive learning, outperforming structure-based methods in speed and accuracy.

This paper introduces PTM-Mamba, a protein language model that incorporates PTM tokens and Mamba blocks to improve performance on PTM-specific tasks.

EvoFlow-RNA generates and represents non-coding RNA using a masked discrete diffusion model, outperforming existing models on RNA function tasks.

moPPIt generates motif-specific and functionally active peptide binders using sequence data, validated in vitro and in disease-relevant assays.

MetaLATTE predicts metal binding in proteins using multi-task learning on ESM-2 embeddings, enabling classification of metal-binding properties without requiring structural input.

This paper introduces FusOn-pLM, a protein language model fine-tuned on fusion oncoprotein sequences using a novel masked language modeling strategy, improving performance on specific benchmarks and disorder prediction tasks.

This paper establishes standardized nomenclature and reporting for analyzing rAAV vectors using PacBio HiFi sequencing to improve quality assessment and vector design.

AMP-Diffusion generates functional antimicrobial peptides using a latent diffusion model integrated with the ESM-2 protein language model, showing alignment with experimentally validated AMPs in pseudo-perplexity and physicochemical properties.

This study shows that co-culturing human oocytes with ovarian support cells improves in vitro maturation rates and blastocyst formation compared to standard IVM methods.

This study shows that co-culturing immature human oocytes with hiPSC-derived ovarian support cells improves in vitro maturation rates and transcriptomic similarity to in vivo matured oocytes.

This study generates functional ovarian granulosa-like cells from human iPSCs by overexpressing NR5A1 and RUNX1 or RUNX2, enabling the formation of ovary-like organoids with germ cells.

This study identifies key transcription factors that drive the rapid differentiation of human induced pluripotent stem cells into oogonia-like cells using a combinatorial approach and computational tools.

This paper designs peptides to degrade undruggable proteins using contrastive deep learning and experimental validation in human cells.

The paper introduces a graph transformer architecture that uses MSA embeddings to predict protein structures directly from sequences.