XtalPi Inc.

𝐗𝐭𝐚𝐥𝐏𝐢 𝐎𝐩𝐞𝐧-𝐒𝐨𝐮𝐫𝐜𝐞𝐬 𝐭𝐡𝐞 𝐅𝐢𝐫𝐬𝐭 𝐀𝐈 𝐌𝐨𝐝𝐞𝐥 𝐟𝐨𝐫 𝐂𝐡𝐞𝐦𝐢𝐜𝐚𝐥 𝐑𝐞𝐚𝐜𝐭𝐢𝐨𝐧 𝐘𝐢𝐞𝐥𝐝 𝐐𝐮𝐚𝐧𝐭𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧 What if chemists could measure reaction yields without preparing a pure reference standard for every new compound? In high-throughput experimentation, researchers run many reactions to find what works. Yet measuring the yield of each reaction often requires pure reference standards, adding time and effort to the process. To address this, XtalPi has developed and 𝐧𝐨𝐰 𝐨𝐩𝐞𝐧-𝐬𝐨𝐮𝐫𝐜𝐞𝐝 𝐋𝐂𝐌𝐒2𝐘𝐢𝐞𝐥𝐝, 𝐚𝐧 𝐀𝐈 𝐦𝐨𝐝𝐞𝐥 𝐭𝐡𝐚𝐭 𝐡𝐞𝐥𝐩𝐬 𝐪𝐮𝐚𝐧𝐭𝐢𝐟𝐲 𝐫𝐞𝐚𝐜𝐭𝐢𝐨𝐧 𝐲𝐢𝐞𝐥𝐝𝐬 without requiring a pure reference standard for each target compound. Here's how it works: 🔹 AI predicts UV correction factors directly from molecular structures. 🔹 These predictions are combined with LC-UV measurements and a known internal standard to calculate product concentrations and reaction yields. 🔹 Across 178 reactions, the method 𝐚𝐜𝐡𝐢𝐞𝐯𝐞𝐝 𝐚 4.74% 𝐦𝐞𝐚𝐧 𝐚𝐛𝐬𝐨𝐥𝐮𝐭𝐞 𝐲𝐢𝐞𝐥𝐝 𝐞𝐫𝐫𝐨𝐫 𝐜𝐨𝐦𝐩𝐚𝐫𝐞𝐝 𝐰𝐢𝐭𝐡 𝐜𝐨𝐧𝐯𝐞𝐧𝐭𝐢𝐨𝐧𝐚𝐥 𝐦𝐞𝐭𝐡𝐨𝐝𝐬. 🔹The model is 10⁵–10⁶ 𝐭𝐢𝐦𝐞𝐬 faster than traditional quantum chemistry calculations. Now available on GitHub, LCMS2Yield includes model weights, code, and datasets, allowing researchers to test the model, reproduce results, and build on the work. If you find this useful, give us a ⭐ on GitHub. 🔗 Explore the open-source model: https://lnkd.in/e8yhegyJ 📄 Read the research in JACS Au: https://lnkd.in/e4RSVw4Q #XtalPi #AIforScience #OpenSource #Chemistry #HighThroughputExperimentation

XtalPi Nominates First Preclinical Candidate from Kodexia™ siRNA Platform in 9 Months The candidate is developed for IgA nephropathy, showed IC50 values tens of times lower than those of clinical-stage reference molecules in lab tests. In non-human primates, a single dose kept its target protein suppressed for months. And there's more. 𝗞𝗼𝗱𝗲𝘅𝗶𝗮'𝘀 𝗔𝗜 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗲𝗱 𝘄𝗵𝗶𝗰𝗵 𝗺𝗼𝗹𝗲𝗰𝘂𝗹𝗲𝘀 𝘄𝗼𝘂𝗹𝗱 𝗽𝗲𝗿𝗳𝗼𝗿𝗺 𝗯𝗲𝘁𝘁𝗲𝗿, 𝗮𝗻𝗱 𝘁𝗵𝗼𝘀𝗲 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻𝘀 𝗰𝗹𝗼𝘀𝗲𝗹𝘆 𝗺𝗮𝘁𝗰𝗵𝗲𝗱 𝘁𝗵𝗲 𝗮𝗻𝗶𝗺𝗮𝗹 𝘀𝘁𝘂𝗱𝘆 𝗿𝗲𝘀𝘂𝗹𝘁𝘀. This is part of a bigger story. By connecting generative AI with automated experiments and real lab data, Kodexia™ helps scientists design, test, and improve siRNA molecules faster. The platform now supports six drug discovery programs across multiple disease areas. We're working to make AI-guided drug discovery faster, more reliable, and repeatable across more targets and diseases. #AI #AIDD #Platform #Medicine #DrugDiscovery #siRNA #Biotech #Kidney

Get the latest jobs and industry news

Not every target plays nice with small molecules. Protein–protein interactions (PPIs) in particular remain elusive, due to their large flat surface area. Peptides excel at this challenge. Their protein structure allows them to interface with PPIs, bridging the gap between small molecules and biologics. And when it comes to identifying the perfect hit, AI is the perfect partner: from backbone to sequence to screening, PepiX brings together multiple models to rationally design peptides for any target.

One week to SWITCH | Singapore Week of Innovation and Technology 2026! As AI moves from prediction to real-world discovery, we're bringing together leaders and innovators to explore what's next for science and technology. Join XtalPi for our AI for Science event in Singapore! We'll welcome representatives from the Singapore government, Stephen Sunderland, Head of APAC at 艾意凯咨询公司, representatives from NVIDIA Healthcare, and Dr. Yalong Zhao, Co-founder of infevo-ai. From XtalPi, our Co-Founder & CIO, Dr. Lipeng Lai, and senior scientists Honggen Zhang, Genwei Zhang, and Lingyu Li will share the latest advances in AI-driven scientific discovery. Here's what to expect: 🔹 Discover how AI models, scientific agents, and automated labs are transforming scientific research. 🔹 Explore advances in AI-driven drug discovery, from next-generation therapeutics to connected research workflows. 🔹 Exchange ideas and connect with innovators across science, technology, and industry. Looking forward to an afternoon of fresh perspectives and meaningful connections! 📍 Marina Bay Sands Expo, Level 3, Cassia Main 1, Singapore 📅 October 28 | 3:30–6:00 PM #SWITCH2026 #AIforScience #AI #DrugDiscovery #XtalPi

We're excited to take a stage on SWITCH | Singapore Week of Innovation and Technology, a global innovation event organized by Singapore government agency! This year, two XtalPi leaders will join the event and discuss how AI is advancing science. 𝗢𝗰𝘁𝗼𝗯𝗲𝗿 𝟮𝟳 | 𝗗𝗿𝘂𝗴 𝗗𝗶𝘀𝗰𝗼𝘃𝗲𝗿𝘆 & 𝗧𝗲𝗰𝗵𝗕𝗶𝗼 𝗗𝗿. Lipeng Lai, 𝗖𝗼-𝗳𝗼𝘂𝗻𝗱𝗲𝗿 & 𝗖𝗜𝗢 𝗼𝗳 𝗫𝘁𝗮𝗹𝗣𝗶, will join industry leaders to discuss how AI and TechBio are reshaping drug discovery and creating new opportunities across the global innovation landscape. 🎤 𝘍𝘳𝘰𝘮 𝘔𝘰𝘭𝘦𝘤𝘶𝘭𝘦 𝘵𝘰 𝘔𝘢𝘳𝘬𝘦𝘵 – 𝘈𝘐, 𝘛𝘦𝘤𝘩𝘉𝘪𝘰, 𝘢𝘯𝘥 𝘵𝘩𝘦 𝘕𝘦𝘸 𝘎𝘦𝘰𝘨𝘳𝘢𝘱𝘩𝘪𝘦𝘴 𝘰𝘧 𝘋𝘳𝘶𝘨 𝘋𝘪𝘴𝘤𝘰𝘷𝘦𝘳𝘺 📍 Sands Expo, L1 Beyond Stage | 11:20 AM–12:05 PM (SGT) 𝗢𝗰𝘁𝗼𝗯𝗲𝗿 𝟮𝟴 | Materials.AI 𝗗𝗿. 张佩宇, 𝗖𝗦𝗢 𝗼𝗳 𝗫𝘁𝗮𝗹𝗣𝗶, will join an expert panel to explore the next inflection point for Materials AI and what it means for the future of materials discovery. 🎤 𝘛𝘩𝘦 𝘍𝘶𝘵𝘶𝘳𝘦 𝘋𝘦𝘱𝘦𝘯𝘥𝘴 𝘰𝘯 𝘔𝘢𝘵𝘦𝘳𝘪𝘢𝘭𝘴: 𝘞𝘩𝘢𝘵 𝘞𝘪𝘭𝘭 𝘉𝘦 𝘵𝘩𝘦 𝘐𝘯𝘧𝘭𝘦𝘤𝘵𝘪𝘰𝘯 𝘗𝘰𝘪𝘯𝘵 𝘧𝘰𝘳 Materials.AI? 📍Sands Expo, L1 Main Stage | 12:10–12:45 PM (SGT) ✨ 𝗖𝗼𝗻𝘁𝗶𝗻𝘂𝗲 𝘁𝗵𝗲 𝗖𝗼𝗻𝘃𝗲𝗿𝘀𝗮𝘁𝗶𝗼𝗻 𝗮𝘁 𝗫𝘁𝗮𝗹𝗣𝗶'𝘀 𝗣𝗮𝗿𝘁𝗻𝗲𝗿 𝗘𝘃𝗲𝗻𝘁! We're also hosting "𝗔𝗜 𝗳𝗼𝗿 𝗦𝗰𝗶𝗲𝗻𝗰𝗲: 𝗙𝗿𝗼𝗺 𝗣𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻 𝘁𝗼 𝗗𝗶𝘀𝗰𝗼𝘃𝗲𝗿𝘆" on October 28, bringing together leaders from science, industry, and investment to discuss how scientific AI, intelligent agents, and robotic laboratories are advancing drug discovery. 📅 October 28 | 15:30–18:00 (SGT) 📍 Sands Expo, Level 3, Cassia Main 1 Interested in joining us? Register here: https://lnkd.in/e4a4B7hX We look forward to connecting with you in Singapore! #SWITCH2026 #XtalPi #AIForScience #DrugDiscovery #MaterialsAI #TechBio