BioMoDes Newsletter -- Issue #02
Here are the latest tools (from last week) for biomolecular modeling and design.
1. Intro
Welcome to the 2nd Issue of the BioMoDes Newsletter and Top Reads, a newsletter where I share with you the latest tools and my top-reads in Biomolecular Modeling and Design, focusing on computational structural biology and drug design.
If you are a new subscriber, you can read the first issue of the newsletter here. Thank you for subscribing!
Many new tools came out in the last two weeks and I will be sharing them in this issue. I will also share some of my top reads for the past week.
Ok.
2. New tools over the last couple of weeks
-
Multiflow: A generative model from the Jaakkola lab for protein sequence and structure co-design. Multiflow is an approach that combines the DFM (for sequence design) and FrameFlow (for structure generation) models. Multiflow outperforms RFdiffusion on unconditional designability, with comparable diversity and novelty to existing methods. Read more...
-
HELM-GPT: A model for de novo design of macrocyclic peptides. Read more...
-
DeepGlycanSite: A neural network for predicting carbohydrate binding sites on proteins. Read more...
-
ESM All-Atom: An all-atom multi-scale version of Meta's ESM protein language model. ESM-AA extends the protein/amino acid scale modeling of ESM to atom scale modeling to capture non-protein atoms. Read more...
-
EvoBind2: A method for designing novel linear and cyclic protein-binding peptides using just the amino acid sequence of the target protein as input. EvoBind2, similar to EvoBind (v1), presents as an in silico directed evolution method using a strategy analogous to the "diversify-screen/select(-amplify)" cycle of experimental directed evolution. Read more...
-
ESM3: A new start-up EvolutionaryScale, emerging from Meta FAIR, launches with a preprint introducing ESM3, a multimodal generative protein language model. ESM3 is a third generation foundational ESM model that "reasons" over protein sequences, structures, and functions. Read more...
-
Protpardelle: An all-atom diffusion model for generating protein structures. Protpardelle, developed by Po-Ssu Huang's group, is a method that simultaneously models and designs the backbone, sequence, and side chains of the protein being generated. Read more...
3. Top reads
-
A paper, first preprinted mid last year and now published in Nat. Commun. describes the use of evolutionary coupling models (EVcouplings) for the design of protein variants with huge mutational difference from the wild type but also with multiple improved functional properties. The paper utilized the EVcouplings approach to design variants of the TEM-1 beta-lactamase with large sequence variations from the wild type (up to 84 mutations in a variant!) while demonstrating improved properties. Read more...
-
The RCSB (PDB) introduces a web-based pairwise structure alignment tool that integrates with the PDB portal and allows the use of multiple alignment methods including FATCAT, CE, TM-align, or Smith-Waterman 3D. Read more...
-
I came across a paper that reports a framework for integrating FRET measurements with MD simulations and de novo structure prediction to model nucleic acids. The FRET-assisted modeling approach was used to filter a de novo RNA structure prediction ensemble by screening out models that are incompatible with in vitro FRET data. Read more...
4. Bonus
Bruce Donald (Duke University) recently gave a talk about the CryoEM structures of their designed Abs bound to specific targets and AlphaFold 3 incorrect predictions of the Ab complexes, emphasizing that AF 2 & 3 predictions should be carefully (and correctly!) evaluated before making important biological hypotheses based on them. I also added some comments and perspectives on this. Read more...
That's it for this week. Feel free to reach out by replying to this email if you have any suggestions or would like to have other discussions.
I try my best to make the information on this website as accurate as possible. If you find any errors in the contents of this page or any other page on this website, I would greatly appreciate that you kindly get in touch with me at contact@abeebyekeen.com.
If you are interested in joining my free weekly “BioModes and Top Reads” newsletter, please subscribe below.