Knowledge Unlatched’s Post

#OpenScience and Open Data represent key principles for ensuring unrestricted public access to scientific research, enhancing validation, reproducibility, and interdisciplinary collaboration, as exemplified by the free-to-access platforms, such as Zenodo https://ow.ly/MPnT50TPBRP

#OpenScience and Open Data represent key principles for ensuring unrestricted public access to scientific research, enhancing validation, reproducibility, and interdisciplinary collaboration, as exemplified by the free-to-access platforms, such as Zenodo https://ow.ly/TOuK50TPBSi

Though the positive impact of #OpenScience practices on innovation, collaboration and transparency is broadly recognized, their implementation continues to face challenges, as their costs and benefits can be context dependent https://ow.ly/Bpfy50TNPFO

Though the positive impact of #OpenScience practices on innovation, collaboration and transparency is broadly recognized, their implementation continues to face challenges, as their costs and benefits can be context dependent https://ow.ly/A7kL50TNPBC

Text2Protein: A Generative Model for Designated Protein Design on Given Description ABSTRACT Designing protein structures from text is challenging in computational biology. We propose Text2Protein, a pipeline combining large language models (LLMs) with diffusion models to generate full-atomic protein structures from text. Using a conditional diffusion model and the Vicuna-7B language model, we learn data distributions of 6D interresidue coordinates, refined into full-atomic structures with PyRosetta. Trained on a curated RCSB-PDB dataset, Text2Protein focuses on single-chain proteins with 40-256 residues. Our extensive experiments validate Text2Protein’s effectiveness by generating high-fidelity protein structures similar to ground truth proteins using raw texts. We evaluate Text2Protein using multiple metrics, including Mean Square Error (MSE) of 6D coordinates, Rosetta Energy Units (REU), and TM-score. Our results show that 5% of the generated proteins have a TM-score greater than 0.5, indicating similar folds in SCOP/CATH. Additionally, 16% of pairs have a TM-score greater than 0.4, 89% have a TM-score greater than 0.3, and none have a TM-score less than 0.17, below the threshold for unrelated proteins. Text2Protein presents a promising framework for automated protein design, potentially accelerating novel protein discovery. This work opens new avenues for integrating natural language understanding with protein structure generation, with implications in drug discovery, enzyme engineering, and material sciences. PAPER: https://lnkd.in/dRMfAS_Q

How we enable open access really matters. It matters to the lives and work of countless scholars across the world. And it matters to all of us - we benefit daily from outcomes based on research and scholarly outputs. Thanks to Jeffrey Brainard @ Science Magazine, AAAS for taking the time to talk to so many stakeholders. This resulting article is filled with facts and data, but is also grounded in the human experience of researchers and the struggles that some face. Proud to see OASPA's work on re-thinking OA models included as part of the story: https://lnkd.in/guubS3Ns

Though the positive impact of #OpenScience practices on innovation, collaboration and transparency is broadly recognized, their implementation continues to face challenges, as their costs and benefits can be context dependent https://ow.ly/xPcV50TNPCY

Current incentives make quantity the enemy of quality, says Professor Philipp Koellinger, chief scientific officer and co-founder of open access science platform DeSci Labs https://lnkd.in/eqirW-C4

Current incentives make quantity the enemy of quality, says Professor Philipp Koellinger, chief scientific officer and co-founder of open access science platform DeSci Labs https://lnkd.in/eqirW-C4

📢 Read this week’s #OpenPharma digest here! Or sign up to our blog to receive our weekly #OpenScience news round-up and more straight to your inbox! #OpenAccess Public Knowledge Project #PeerReview #FAIRData Center for Open Science #ScholarlyPublishing #Referencing Invest in Open Infrastructure #OpenInfrastructure #InfraFinder