Transformer-based methods for calling 5mC methylations. Increase in the single-base accuracy of up to 5% on read level. W/ Dominik Stanojevic, Zhe Li and Roger Foo biorxiv.org/content/10.110…

Long reads metagenome benchmark is out. Highlights. 1. In most cases Kraken, 2. minimap2/ram for slightly higher accuracy. 3. The right database is of huge importance 4. Check taxonomy files carefully. bmcbioinformatics.biomedcentral.com/articles/10.11… w/Niranjan Nagarajan Krešimir Križanović @jmaricb Sylvain

Are nanopore UL reads only long reads we need? We developed Herro github.com/lbcb-sci/herro AI error correction model that can correct reads to accuracy above Q30 while trying to keep informative positions intact. w/ Dominik Stanojevic Dehui Lin Sergey Nurk 🇺🇦 Pore_XX_Singapore Oxford Nanopore

Happy to present an initial draft of a telomere-to-telomere diploid Indian genome. A joint effort of Jianjun Liu's and my lab spearheaded by Prasadms and Josipa Lipovac github.com/lbcb-sci/I002C or github.com/LHG-GG/I002C

Happy to present RiNALMO - our RNA large language model arxiv.org/abs/2403.00043 w/ Rafael Josip Penić Tin Vlasic Yue Wan and Roland Huber. RiNALMo is the largest RNA language model to date, with 650 million parameters pre-trained on 36 million non-coding RNA sequences. 1/2

Our new paper is out! We made a lot of progress on our GNN-based de novo genome assembly paradigm and here we present all our findings and progress 🧬🥳 Code: github.com/lbcb-sci/GNNome Paper: biorxiv.org/content/10.110… See details of GNNome below 👇🧵

Considering building a human pangenome? Dive into our latest preprint for insights on the sequencing technologies and the minimal coverages needed for accurate assemblies biorxiv.org/content/10.110… Joint work with Prof JJ Liu lab w/Prasadms, Josipa Lipovac and Filip Tomas!!

Rockfish: A transformer-based model for accurate 5-methylcytosine prediction from nanopore sequencing has been published in Nat. Comm!! Great work by Dominik Stanojevic w/ Zhe Li, Sara Bakić and Roger Foo Paper: nature.com/articles/s4146… Code: github.com/lbcb-sci/rockf…

Our RNA LLM RiNALMo has received the prize for the most ambitious submission at the Machine Learning for Life and Material Science workshop at ICML Conference 2024. w/ Rafael Josip Penić Tin Vlasic Yue Wan and Roland Huber. Workshop: ml4lms.bio Preprint: arxiv.org/abs/2403.00043

CoPRA: Bridging Cross-domain Pretrained Sequence Models with Complex Structures for Protein-RNA Binding Affinity Prediction - Curate large datasets of RNA-Protein pairs (PRI30K, 150K pairs) from BioLIP2, and an affinity dataset (PRA310, 435 complexes) from PDBBind, ProNAB, and

We are seeking talented PhD candidates in CS, physics or math to join us. The topics include the development of AI for cancer and ageing research. 📅 Application Deadline: 1st Dec 2024 Lab: sikic-lab.github.io Scholarship: a-star.edu.sg/Scholarships/f… Please share!

A Comparative Review of Deep Learning Methods for RNA Tertiary Structure Prediction biorxiv.org/cgi/content/sh… #biorxiv_bioinfo

RNA structure prediction is still an open problem. Look at our benchmark results (including Alphafold 3!) biorxiv.org/content/10.110… w/ Ivona Martinović Tin Vlasic Yang Li Bryan Hooi and Zhang Yang

🚀Thrilled to be part of ICLR 2025! Join our workshop AI for Nucleic Acids (AI4NA) to explore cutting-edge research and connect with field leaders. Thanks to our organizers and ICLR 2026 for making this possible. Find the link to the website in our bio! More info below👇

🚨 1 month to go! 🚨 The submission deadline for the AI4NA workshop at ICLR 2026 is fast approaching! 🧬 ✨ Submissions on OpenReview will open soon—stay tuned! ✨ 🔗 Learn more on our web page (link below 👇) #AI4NA #ICLR2025

Submit your papers to our workshop AI for Nucleic Acids workshop at ICLR 2025!

We are organising ai4na-workshop.github.io at#ICLR2025. Topics: RNA, DNA and cell LLMs, structure, modifications, correction, variant calling... Deadline: Feb 10th 2025 Time for revision: 1 month Accepted papers - the opportunity to be invited by Nature Methods for submission

I am happy to share our new preprint introducing MADRe - a pipeline for Metagenomic Assembly-Driven Database Reduction, enabling accurate and computationally efficient strain-level metagenomic classification. Mile Sikic, Riccardo Vicedomini, Krešimir Križanović 🔗biorxiv.org/content/10.110… 1/9

I am happy to introduce Campolina, a deep neural framework that replaces traditional algorithmic approaches for nanopore signal segmentation and improves segmentation quality for real-time analysis. Preprint and details in the thread👇 K. Friganovic, Bryan Hooi, Mile Sikic 1/7