nabil_ibtehaz

computerphile

[Kihara Laboratory @kiharalab]

Poster presentations given today at the Merck–Purdue Symposium! Thank you for the very engaging discussions we had with audience. Presentations by Yuki Kagaya, Farhanaz Farheen, and Anika Jain.

[bioRxiv Bioinfo @biorxiv_bioinfo]

Accurate Macromolecular Complex Modeling for Cryo-EM with CryoZeta https://t.co/4J38NQmOZr #biorxiv_bioinfo

[Kihara Laboratory @kiharalab]

New paper:🧬 Queryome: A multi-agent AI system for biomedical literature analysis.
Queryome is a deep research system with specialized LLM agents that orchestrate to a wide range of queries based on Pubmed citations.
https://t.co/jUIukMjUOb
Try it: https://t.co/MGN3wb1F45

[Kihara Laboratory @kiharalab]

New paper released! "Distance-AF improves predicted protein structure models by AlphaFold2 with user-specified distance constraints" Yuanyuan Zhang, Zicong Zhang, Y Kagaya, G Terashi, B Zhao, Y Xiong & D Kihara, Communications Biology. @CommsBio https://t.co/FD8CCP2VOJ

[nabil_ibtehaz @NIbtehaz]

our mayor 🤗🤗🤗

[Polling USA @USA_Polling]

Zohran Mamdani wins the first round of the NYC Democratic mayoral election

Given his lead, we're almost guaranteed to see Mamdani win on the final ballot

[Kihara Laboratory @kiharalab]

New paper online : Learning with Privileged Knowledge Distillation for Improved Peptide–Protein Docking, ACS Omega. The method, Distpepfold, uses student-teacher model to improve peptide docking over AF2.
https://t.co/Mbo0FmzZ4C

[Kihara Laboratory @kiharalab]

Flash talk by Yuki Kagaya on the Nufold RNA structure prediction method presented at the CASP16 evaluation meeting:
https://t.co/knGTKYpJUQ

You can run Nufold at Google Colab: https://t.co/K5YAidIMMD
Nufold paper: https://t.co/3Q2f2irx2L

[Kihara Laboratory @kiharalab]

Pictures from the CS graduate symposium last week. Swagarika Giri talked about her work on Go2Sum, protein function summarizer, and Daisuke Kihara was on a faculty panel discussion. https://t.co/vagEhLkZea

[Kihara Laboratory @kiharalab]

Professor Kihara receives 2025 College of Science Faculty Research Award
https://t.co/knXEW7wrqS

[Biology+AI Daily @BiologyAIDaily]

NuFold: end-to-end approach for RNA tertiary structure prediction with flexible nucleobase center representation @NatureComms

1. NuFold is a novel deep learning-based framework that predicts RNA tertiary structures directly from sequence data, offering a breakthrough in RNA structure prediction by enabling fully atomic models without reliance on external templates.

2. A key innovation of NuFold is its flexible nucleobase center representation, which optimizes RNA backbone flexibility, capturing important conformational variations like sugar-pucker states and enhancing the precision of structure prediction.

3. Unlike previous RNA structure prediction methods, NuFold operates in an end-to-end fashion, eliminating the need for intermediate distance or angle constraints, which streamlines the modeling process and improves overall accuracy.

4. In comparative tests, NuFold demonstrated superior performance over energy-minimization methods like SimRNA and FARFAR2, achieving an RMSD of 5 Å or less for many RNA structures, including tRNAs and riboswitches, indicating high structural fidelity.

5. The integration of metagenomic sequences for multiple sequence alignment (MSA) and the use of recycling during prediction significantly improved model accuracy, underscoring the importance of dataset diversity and iterative refinement.

6. Through case studies, NuFold proved its ability to model complex RNA structures, such as riboswitches and ribozymes, with remarkable precision, including some challenging targets from RNA-Puzzles.

7. Future developments will explore multi-chain RNA structures and potential drug-target RNA designs, highlighting NuFold's applicability in broader biological and therapeutic contexts.

@d_kihara @kiharalab @NIbtehaz
💻Code: https://t.co/dehWLG1cXV
📜Paper: https://t.co/Fd2EBXnq5G

#RNAstructure #DeepLearning #NuFold #ComputationalBiology #Bioinformatics #MachineLearning #DrugDiscovery

[Kihara Laboratory @kiharalab]

"NuFold: end-to-end approach for RNA tertiary structure prediction with flexible nucleobase center representation" by Yuki Kagaya et al. is now out in @NatureComms !
This deep learning method played a key role in our 3rd place finish for RNA in CASP16.
https://t.co/3Q2f2irx2L

[nabil_ibtehaz @NIbtehaz]

Read our paper : https://t.co/pUhiwUeOpq
We also have a google colab server for predicting RNA structures using NuFold : https://t.co/LFzgK58mVN
Our codes are open-source : https://t.co/VhqbvKUGSo

[nabil_ibtehaz @NIbtehaz]

I am excited to share that our paper "NuFold: end-to-end approach for RNA tertiary structure prediction with flexible nucleobase center representation" has been published in Nature Communications #rna #rna_3d_structure #bioinformatics🥳🥳🥳📷

[nabil_ibtehaz @NIbtehaz]

NuFold showcased exceptional performance in the recent CASP16 competition,contributing to our lab achieving 3rd place in RNA structure prediction.