Orange Data Mining’s Post

🧬 What if we could design proteins that nature never created? AI is changing the way we approach biology — from predicting protein structures to exploring entirely new protein sequences. But the real breakthrough isn't AI alone. It happens when AI + bioinformatics + molecular biology + experimental science come together. 🔬🤖 At LabbeyondX, we want to explore these intersections, learn the tools behind them, and turn curiosity into real scientific skills and projects. 🌍 Science has no borders. What would you design if you had the tools to create a new protein? #LabbeyondX #AIinBiotech #ProteinDesign #Biotechnology #Bioinformatics #ComputationalBiology #SyntheticBiology #MolecularBiology #LifeSciences #ScienceCommunity

Developed by OpenAI, Rosalind Workbench unifies biological research data, analysis, and scientific tools into a single workflow. While general-purpose AI models can present information, life science research demands multi-step workflows bridging biological data, scientific tools, and literature. 💻 Rosalind Workbench addresses this fragmented structure within a more integrated research environment. At its core, GPT-Rosalind combines domain-specific reasoning in life sciences with scientific tool utilization and evidence synthesis. 🧬 The system aims to facilitate a seamless transition from inquiry to data, and from data to scientific insight, across domains such as genomics, sequencing, protein analysis, and variant analysis. 🔬 Consequently, AI evolves from an information-retrieval system into a functional assistant capable of interfacing with tools across various stages of scientific inquiry. #rosalindworkbench #bioinformatics #genomics #ngs #bioinfocodes

🚀 Had a fantastic time at the European Conference on Computational Biology (ECCB) presenting my poster on BI-SCUDO Regression! 🚀 It was an honor to share our research with the international bioinformatics community. My poster, "BI-SCUDO Regression: Robust Explainable Regression for Noisy Omics via a Rank-Based Algorithm," addresses the crucial challenge of making sense of noisy, high-dimensional omics data. Our method, BI-SCUDO, leverages a rank-based algorithm to achieve: ✅ Robustness against outliers and noise common in biological data. ✅ Interpretability of the results, providing explainable regression models. I was thrilled by the engagement and the many insightful questions during the session. It was inspiring to see so much incredible work happening in computational biology and to connect with talented colleagues. Special thanks to my co-authors Mario Lauria, Roberto Bizzotto and Carlo Alberto Rossi and my PhD supervisor Luca Marchetti for their collaboration and help and the Università di Trento and UniTrento CIBIO for making this research possible. #ECCB #Bioinformatics #ComputationalBiology #Omics #DataScience #ExplainableAI #MachineLearning #UniversityofTrento

🧬 DNA JUST GOT A LOT MORE INTERESTING! For billions of years, life on Earth has relied on a simple four-letter genetic alphabet: A, T, C and G. But scientists have now taken an exciting step toward expanding that alphabet to eight letters! 🔬 Researchers demonstrated that E. coli RNA polymerase—the enzyme responsible for copying DNA into RNA—can recognize and transcribe four additional synthetic DNA letters: P, Z, B and S. These form two new base pairs alongside the familiar A–T and C–G pairs. Using advanced cryo-electron microscopy, researchers were able to see how the synthetic bases fit into the enzyme and discovered that the molecular machinery could recognize them using mechanisms similar to those used for natural DNA bases. This doesn't mean scientists have created an eight-letter organism yet. There are still major challenges, particularly getting the expanded genetic information to work through the entire DNA → RNA → protein pathway. However, the discovery opens exciting possibilities for synthetic biology, new molecular tools, diagnostics, medicines and engineered biological systems. The big idea is fascinating: nature uses four genetic letters, but biology's molecular machinery may be capable of reading a much larger genetic vocabulary.🧬🌍 Science keeps reminding us that there is still so much to discover about the language of life. The Brighter Side of News. (2026). Biology’s four-letter DNA alphabet just expanded to eight. The Brighter Side of News. #Biology #Genetics #DNA #MolecularBiology #SyntheticBiology #Biotechnology #Science #Genomics

At RAIS Code, we bring dedicated bioinformatics expertise to research teams — without the overhead of building an in-house team. Our services span three core areas: 🧬 Genomics & Sequence Analysis — whole genome analysis, variant annotation & mutational analysis, primer design 💊 Drug & Vaccine Design — in silico modeling and computational design 🌱 Environmental & Applied Bioinformatics — in silico biodegradation studies of pollutants using microorganisms Whether you need a single dataset analyzed or an ongoing computational partner for your lab, we'd love to talk about how we can support your research. 📩 Reach out to learn more. #Bioinformatics #Genomics #ComputationalBiology #DrugDesign #VaccineDesign #ScientificComputing #RAISCode

Modern biology is generating more data than ever, from genomic and transcriptomic datasets to molecular and clinical data. 𝐁𝐢𝐨𝐢𝐧𝐟𝐨𝐫𝐦𝐚𝐭𝐢𝐜𝐬 𝐡𝐞𝐥𝐩𝐬 𝐭𝐮𝐫𝐧 𝐭𝐡𝐢𝐬 𝐝𝐚𝐭𝐚 𝐢𝐧𝐭𝐨 𝐦𝐞𝐚𝐧𝐢𝐧𝐠𝐟𝐮𝐥 𝐛𝐢𝐨𝐥𝐨𝐠𝐢𝐜𝐚𝐥 𝐢𝐧𝐬𝐢𝐠𝐡𝐭𝐬. Learning bioinformatics gives you the computational foundation to work with biological datasets, understand patterns in complex data, perform analysis, and support research-driven discoveries. Whether you're studying genomics, transcriptomics, drug discovery, precision medicine, or computational biology, these skills are becoming increasingly relevant across modern life science research. 🔬 Start with the fundamentals and gradually build your understanding through structured courses, practical analysis, and real biological datasets. 📚 𝐄𝐱𝐩𝐥𝐨𝐫𝐞 𝐭𝐡𝐞 𝐂𝐨𝐮𝐫𝐬𝐞 𝐋𝐢𝐛𝐫𝐚𝐫𝐲: https://lnkd.in/gAs2sahm 🎓 𝐕𝐢𝐞𝐰 𝐒𝐮𝐛𝐬𝐜𝐫𝐢𝐩𝐭𝐢𝐨𝐧 𝐏𝐥𝐚𝐧𝐬: https://lnkd.in/ghhyqhXy 𝐁𝐮𝐢𝐥𝐝 𝐭𝐡𝐞 𝐬𝐤𝐢𝐥𝐥𝐬 𝐭𝐨 𝐰𝐨𝐫𝐤 𝐰𝐢𝐭𝐡 𝐭𝐡𝐞 𝐝𝐚𝐭𝐚 𝐝𝐫𝐢𝐯𝐢𝐧𝐠 𝐦𝐨𝐝𝐞𝐫𝐧 𝐛𝐢𝐨𝐥𝐨𝐠𝐲. 🧬 #Bioinformatics #ComputationalBiology #Genomics #Transcriptomics #LifeSciences #BiologicalData #BioinformaticsLearning #OmicsLogic

Modern biology is generating more data than ever, from genomic and transcriptomic datasets to molecular and clinical data. 𝐁𝐢𝐨𝐢𝐧𝐟𝐨𝐫𝐦𝐚𝐭𝐢𝐜𝐬 𝐡𝐞𝐥𝐩𝐬 𝐭𝐮𝐫𝐧 𝐭𝐡𝐢𝐬 𝐝𝐚𝐭𝐚 𝐢𝐧𝐭𝐨 𝐦𝐞𝐚𝐧𝐢𝐧𝐠𝐟𝐮𝐥 𝐛𝐢𝐨𝐥𝐨𝐠𝐢𝐜𝐚𝐥 𝐢𝐧𝐬𝐢𝐠𝐡𝐭𝐬. Learning bioinformatics gives you the computational foundation to work with biological datasets, understand patterns in complex data, perform analysis, and support research-driven discoveries. Whether you're studying genomics, transcriptomics, drug discovery, precision medicine, or computational biology, these skills are becoming increasingly relevant across modern life science research. 🔬 Start with the fundamentals and gradually build your understanding through structured courses, practical analysis, and real biological datasets. 📚 𝐄𝐱𝐩𝐥𝐨𝐫𝐞 𝐭𝐡𝐞 𝐂𝐨𝐮𝐫𝐬𝐞 𝐋𝐢𝐛𝐫𝐚𝐫𝐲: https://lnkd.in/gAs2sahm 🎓 𝐕𝐢𝐞𝐰 𝐒𝐮𝐛𝐬𝐜𝐫𝐢𝐩𝐭𝐢𝐨𝐧 𝐏𝐥𝐚𝐧𝐬: https://lnkd.in/ghhyqhXy 𝐁𝐮𝐢𝐥𝐝 𝐭𝐡𝐞 𝐬𝐤𝐢𝐥𝐥𝐬 𝐭𝐨 𝐰𝐨𝐫𝐤 𝐰𝐢𝐭𝐡 𝐭𝐡𝐞 𝐝𝐚𝐭𝐚 𝐝𝐫𝐢𝐯𝐢𝐧𝐠 𝐦𝐨𝐝𝐞𝐫𝐧 𝐛𝐢𝐨𝐥𝐨𝐠𝐲. 🧬 #Bioinformatics #ComputationalBiology #Genomics #Transcriptomics #LifeSciences #BiologicalData #BioinformaticsLearning #OmicsLogic

Biology’s four-letter DNA alphabet just expanded to eight - The Brighter Side of News https://lnkd.in/dac3umvU 🧬 What if DNA had more than 4 letters? For all the biology we know, DNA has been built around just four letters: A • T • G • C But scientists have now pushed that boundary further. 🔬 Researchers have demonstrated that E. coli RNA polymerase can read and transcribe an eight-letter genetic alphabet called Hachimoji DNA. Alongside the natural base pairs: 🔹 A–T 🔹 G–C the system introduces two synthetic pairs: 🔹 P–Z 🔹 B–S And here’s the fascinating part 👇 The enzyme doesn't simply tolerate these artificial bases—it can recognize and incorporate them using structural mechanisms remarkably similar to those used for natural nucleotides. Using cryo-EM, researchers captured RNA polymerase interacting with these synthetic bases at near-atomic resolution. The structures revealed that the artificial base pairs can adopt familiar Watson–Crick-like geometry inside the enzyme. ⚠️ But this does NOT mean we now have an eight-letter organism. The work demonstrates transcription in a controlled biochemical system. Major challenges—including reliable replication and translation into proteins—still remain. 🌱 Why does this matter? Expanding the genetic alphabet could give synthetic biology a much larger molecular vocabulary. More letters → more possible sequences → potentially more chemical diversity → potentially new molecular functions, diagnostics, therapeutics and engineered biological systems. It makes you wonder: If nature's genetic alphabet isn't the only possible alphabet, how much more can we engineer biology to do? 🧬 This is where synthetic biology, structural biology, bioinformatics and AI-driven molecular design could become incredibly powerful. #DNA #SyntheticBiology #Bioinformatics #Genomics #GeneticEngineering #MolecularBiology #StructuralBiology #CryoEM #Biotechnology #LifeScience #AIinBiology #HachimojiDNA

Back in July, I wrote that I'd spend some time building independent tools for researchers. Project #1 is done. I Built EBV-KG, an open-source knowledge system dedicated to Epstein–Barr virus research. The goal was to go beyond basic paper searches and create a platform that links discoveries across virology, immunology, genetics, single-cell biology, and disease, while also recording the sources of these connections. EBV has always been one of my favorite viruses to study, and it probably still is. I enjoyed revisiting the virus and examining its relationship from a new perspective. The last time I was involved in EBV research was when I was beginning my journey in bioinformatics. The current system includes: 13,000+ biomedical documents 740,000+ extracted candidate relationships a curated knowledge graph semantic literature retrieval multi-hop graph traversal literature provenance .h5ad / single-cell data ingestion FastAPI + interactive graph visualisation There was considerably more engineering involved than I initially expected. Storage, databases, entity normalisation, extraction quality, graph pruning and infrastructure all became part of the problem. I've written up the project, including why I built it, how it works, what broke, and what I learned from applying it to actual EBV biology. Article: Link below in comments Code: https://lnkd.in/e6gf7wR4 This is the first of several research-tooling projects I want to build. #Bioinformatics #ComputationalBiology #ResearcherTooling #KnowledgeGraph #EBV #OpenSource #AIforScience