←
Bin Zhang167 posts
Bin Zhang avatar

Bin Zhang

@binzmit

Associate professor of chemistry, MIT

◷ Joined November 2017▧ 11 Photos and videos
Bin Zhang avatar
Bin Zhang@binzmit·Oct 21

New preprint: CV-SVGD combines Stein variational gradient descent with parallel restrained MD to explore higher-dimensional collective-variable spaces. In peptide tests, it recovered free-energy surfaces with less MD sampling than umbrella sampling. arxiv.org/abs/2610.27429

grabstract avatar
⟳ Bin Zhang retweeted
grabstract@grabstractio·Oct 16

Proteins face a molecular trade-off inside cells: the forces that help them quickly clump together into droplets can actually ruin their internal organization and order. @binzmit doi.org/10.1021/acs.jpcb.6c0…

1:03
Bin Zhang avatar
Bin Zhang@binzmit·Oct 15

Our new paper introduces a molecular dynamics model for limited-valency sticker interactions in biomolecular condensates. Simulations show spacer interactions can promote condensation while disrupting sticker-mediated organization and robust recruitment. doi.org/10.1021/acs.jpcb.6c0…

Bin Zhang avatar
Bin Zhang@binzmit·Oct 15

How open is “open” chromatin? Our new Nature Genetics paper uses nucleosome-resolution modeling to reveal condensed euchromatin domains with short, accessible regulatory regions at the surface. nature.com/articles/s41588-0…

MIT Chemistry avatar
⟳ Bin Zhang retweeted
MIT Chemistry@ChemistryMIT·Jan 12

Congratulations to Professor Bin Zhang, who has been honored as “Committed to Caring”. The C2C program is a student-driven initiative that recognizes outstanding professors who extend this dedication beyond the classroom. chemistry.mit.edu/chemistry-…

Post attachment
Bin Zhang avatar
Bin Zhang@binzmit·Jan 12

Pretraining implicit solvent, or coarse grained, models is difficult, limiting the transferability of the resulting ML force fields. However, Justin just figured out a novel method for pretraining using protein language models. More can be found at: arxiv.org/abs/2601.05388

Bin Zhang avatar
Bin Zhang@binzmit·5 Dec 2025

Thrilled to share our new JCP editorial (shorturl.at/6mKdY), co-authored with @tamar_schlick, introducing the special collection “Chromatin Structure and Dynamics: Recent Advancements.” Huge thanks to all the outstanding contributors!

Bin Zhang avatar
Bin Zhang@binzmit·31 Jan 2025

Our paper is now online at Sci Adv: science.org/doi/10.1126/scia… See also the accompanying news: news.mit.edu/2025/with-gener…

I am excited to share our latest #preprint: ChromoGen: Diffusion model predicts single-cell chromatin conformations researchsquare.com/article/r… As the title suggests, we achieved de novo prediction of 3D chromatin structures using DNA sequence and ATAC-seq using AI.

Bin Zhang avatar
Bin Zhang@binzmit·15 Dec 2024

Hmm, I assume these posts were made by AI? Anyway, very impressive summary of our latest preprint.

Protein Language Model Identifies Disordered, Conserved Motifs Driving Phase Separation 1. This study employs ESM2, a cutting-edge protein language model, to analyze intrinsically disordered regions (IDRs) in proteins, uncovering conserved motifs critical for phase separation and membraneless organelle (MLO) formation. 2. A major finding reveals that IDRs involved in phase separation contain conserved “sticker” residues (e.g., Y, W, F) and “spacer” residues (e.g., G, A, P), forming functional sequence motifs under evolutionary pressure. 3. By predicting mutational constraints, ESM2 accurately identifies conserved, functional residues and motifs without relying on sequence alignments, overcoming a key limitation in studying disordered proteins. 4. Experimental validation shows that many conserved motifs identified by ESM2 are essential for phase separation. Mutations within these motifs disrupt MLO formation, highlighting their biological significance. 5. The study introduces a motif-based framework for understanding the “grammar” of IDRs, emphasizing how conserved motifs integrate structural flexibility with functional specificity. 6. This work bridges computational predictions with experimental biology, advancing our understanding of IDRs in phase separation and offering insights into disease-linked protein dysfunction. 7. ESM2 emerges as a powerful tool for investigating the evolutionary and functional landscapes of disordered proteins, with broad implications for molecular biology and synthetic biology. @binzmit @yumengzhang99 📜Paper: biorxiv.org/content/10.1101/… #ProteinScience #Bioinformatics #PhaseSeparation #MachineLearning #IntrinsicallyDisorderedRegions

Post attachment
Yunrui Qiu avatar
⟳ Bin Zhang retweeted
Yunrui Qiu@YunruiQ·19 Nov 2024

Our findings shed light on the significant regulatory roles of bio condensate and DNA linker length and help bridge the gap between in vivo and in vitro observations.

0:18
Gene Regulation avatar
⟳ Bin Zhang retweeted
Gene Regulation@generegulation·17 Nov 2024

Nucleosome condensate and linker DNA alter chromatin folding pathways and rates [Qiu et al, 2024] biorxiv.org/content/10.1101/… ▶️residue-level coarse-grained models; non-Markovian dynamics ▶️10n bp DNA linker lengths favor zigzag fibrils ▶️10n+5 bp chromatin loses unique conformations

Post attachment
Biology+AI Daily avatar
⟳ Bin Zhang retweeted
Biology+AI Daily@BiologyAIDaily·8 Oct 2024

Scaling Graph Neural Networks to Large Proteins 1. This paper introduces the DISPEF dataset, specifically designed for benchmarking graph neural networks (GNNs) on large, biologically relevant proteins. DISPEF contains over 200,000 proteins with implicit solvation free energies, a key challenge for molecular modeling. 2. A major innovation is the introduction of a multiscale architecture called “Schake,” which enhances GNN performance on large proteins by incorporating both short-range and long-range interactions, ensuring computational efficiency and transferability across protein sizes. 3. The Schake model leverages two types of message-passing layers: a more accurate SAKE layer for short-range atomic environments and a more efficient SchNet layer for long-range alpha carbon interactions. This mixed design significantly improves both accuracy and speed. 4. DISPEF provides a diverse chemical environment, including both folded and disordered protein regions, making it an ideal dataset for testing GNNs. The inclusion of many-body solvation free energies as targets pushes the limits of model accuracy and generalization. 5. Benchmark results show that Schake consistently outperforms existing GNNs on energy and force predictions for large proteins, while maintaining superior computational efficiency, reducing inference times by up to 2.7× compared to prior models. 6. A key insight from the study is that increasing the cutoff distance in GNNs improves model transferability to larger proteins, but this needs to be balanced against computational cost, which Schake addresses with its hybrid architecture. 7. This work highlights the importance of datasets like DISPEF for driving future GNN innovations and optimizing models for real-world applications like protein folding and drug discovery, where large proteins are common. 8. The paper provides valuable insights for advancing GNN architectures in computational biology, emphasizing the need for both accuracy and efficiency in handling the vast complexity of biomolecular systems. @binzmit 💻Code: github.com/ZhangGroup-MITChe… 📜Paper: arxiv.org/abs/2410.03921

Post attachment