ELIXIR urges European policymakers to invest in open biodata, a strategic resource for competitiveness, innovation, and AI development, by strengthening infrastructure and skills in Europe.
Italy's CNR seeks companies for the BEST project, bridging ELIXIR-IT research with industry through Open Science collaboration. Deadline: June 20, 2025.
ELIXIR-IT's All Hands Meeting (Oct 13-14, 2025) features a "Stories, Projects, Perspectives" session. Members can submit abstracts by July 20. Workshop proposals from Platforms, Communities, and Focus Groups are due July 7.
CNR invites companies to submit an Expression of Interest for the SUNRISE project by May 23.
The B1MGplus project, with ELIXIR-IT participating, launched to enable sustainable, secure access for personalized healthcare.
Non-canonical nucleotides are widespread chemical modifications in eukaryotic genomes and transcriptomes, especially within the epitranscriptome. Traditional enrichment-based sequencing methods cannot detect them at single-molecule resolution, a limitation overcome by nanopore sequencing. Researchers from the University of Bari (Italy) present a transformer-based model for the ab initio detection of multiple RNA modifications, along with two tools, NanoListener and NanoSpeech, to create training datasets and develop modification-aware basecallers. Their approach enables the simultaneous identification of up to nine ribonucleotide variants.
Protein embeddings encode sequence information using protein language models. The Embedding-Based Alignment (EBA) method enables pairwise embedding comparison and remote homolog detection. In this paper we show that, when applied to a set of multifunctional human proteins, EBA identifies remote homologs sharing structural and functional features, validated on the Swiss-Prot database.
CREDO, a Customizable, REproducible, DOcker file generator for bioinformatics applications, has been developed as a tool to moderate reproducibility issues by building and distributing docker containers with embedded bioinformatics tools. CREDO simplifies the process of generating Docker images, facilitating reproducibility and efficient research in bioinformatics.
In this work, it is evaluated the possibility of improving AlphaFold2’s three-dimensional predictions by developing a new pipeline (AlphaMod) that integrates AlphaFold2 with MODELLER, a traditional homology modeling program. For this study, MARCONI100 at HPC@CINECA was used, which was made available through an ELIXIR project.