EPFL School of Life Sciences’ Post

Ready to push the boundaries of science? Enveda is offering two exciting challenges for the Bio x ML hackathon we're co-hosting with our friends at Lux Capital and EvolutionaryScale: 1️⃣ Enhancing MS/MS Fragmentation Predictions: Leverage our GrAFF-MS model, which uses graph neural networks to predict mass spectrometry fragmentation patterns, or develop your own solution. You'll have access to ~1 million curated MS/MS spectra. 2️⃣ Advancing Molecular Similarity: Develop models that classify molecular similarity, capturing the nuanced perspective of expert chemists. Train your model on larger datasets and fine-tune it with Enveda's curated set of 600 molecular pairs, each annotated by five chemists. 💰 Prize: $10,000 for the best team, plus the opportunity to present your work to, and work with, Enveda’s team. 🌐 Open to participants worldwide. More info here: https://lnkd.in/ek7Z6zRV Join the hackathon and make an impact: https://hackathon.bio/

🔊 Join the next Turing-Roche Translational Science Methods Club (T-SciM Club)! 🗓 23 May 2024 3.30pm-5pm BST The T-SciM Club organised by Sarah Katharina Bühler aims to bring together early career researchers to discuss different methods that help us translate from data to science to clinic and provide attendees with a better understanding of a range of quantitative methods and inferences used for different types of health data science challenges. This month's virtual event is on the theme of Multimodal Data Integration ↘ We will be joined by Yuhan Wang from King’s College London and Hui Xin Ng from UC San Diego. 👉 Yuhan will discuss integrating multimodal data into various medical applications. Using Alzheimer’s disease as a focal point, she will illustrate the integration of image, genetic, and tableau data to detect the disease. 👉 Hui Xin will give a demo on how to use the Python library Captum to understand which brain features from a neuroimaging dataset are most important and how a model she is currently using reached its prediction. ✍ Register here: https://lnkd.in/eT-dH8pJ

🌌 𝗥𝗲𝗳𝗹𝗲𝗰𝘁𝗶𝗻𝗴 𝗼𝗻 𝗮𝗻 𝗜𝗻𝗰𝗿𝗲𝗱𝗶𝗯𝗹𝗲 𝗘𝘅𝗽𝗲𝗿𝗶𝗲𝗻𝗰𝗲 𝗮𝘁 𝘁𝗵𝗲 𝗣𝗵𝘆𝘀𝗶𝗰𝘀 𝗼𝗳 𝗗𝗮𝘁𝗮 𝗪𝗼𝗿𝗸𝘀𝗵𝗼𝗽 🚀 I recently had the privilege of co-organizing the 𝗪𝗼𝗿𝗸𝘀𝗵𝗼𝗽 𝗳𝗼𝗿 𝗣𝗵𝘆𝘀𝗶𝗰𝘀 𝗼𝗳 𝗗𝗮𝘁𝗮, held at the beautiful 𝗔𝘀𝗶𝗮𝗴𝗼 𝗢𝗯𝘀𝗲𝗿𝘃𝗮𝘁𝗼𝗿𝘆 on the 27th and 28th of September. This event brought together experts from academia and industry, exploring the intersection of 𝗽𝗵𝘆𝘀𝗶𝗰𝘀, 𝗱𝗮𝘁𝗮 𝘀𝗰𝗶𝗲𝗻𝗰𝗲, 𝗺𝗮𝗰𝗵𝗶𝗻𝗲 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴, and 𝗾𝘂𝗮𝗻𝘁𝘂𝗺 𝘁𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝗶𝗲𝘀 to address some of the most pressing global challenges we face today. From tackling 𝘤𝘭𝘪𝘮𝘢𝘵𝘦 𝘤𝘩𝘢𝘯𝘨𝘦 and 𝘱𝘶𝘣𝘭𝘪𝘤 𝘩𝘦𝘢𝘭𝘵𝘩 𝘤𝘳𝘪𝘴𝘦𝘴 to advancing the 𝘥𝘪𝘨𝘪𝘵𝘢𝘭 𝘳𝘦𝘷𝘰𝘭𝘶𝘵𝘪𝘰𝘯, it’s clear that the role of 𝘥𝘢𝘵𝘢-𝘥𝘳𝘪𝘷𝘦𝘯 𝘱𝘩𝘺𝘴𝘪𝘤𝘴 is more crucial than ever. The workshop featured fascinating discussions on topics such as: - 𝗦𝗶𝗺𝘂𝗹𝗮𝘁𝗶𝗼𝗻𝘀 𝗳𝗼𝗿 𝗟𝗮𝗿𝗴𝗲 𝗛𝗮𝗱𝗿𝗼𝗻 𝗖𝗼𝗹𝗹𝗶𝗱𝗲𝗿 (𝗟𝗛𝗖) 𝗲𝘃𝗲𝗻𝘁𝘀 and real-time data selection at 𝗖𝗠𝗦 🌀 - 𝗤𝘂𝗮𝗻𝘁𝘂𝗺 𝗰𝗼𝗺𝗽𝘂𝘁𝗶𝗻𝗴 𝗮𝗻𝗱 𝘁𝗲𝗻𝘀𝗼𝗿 𝗻𝗲𝘁𝘄𝗼𝗿𝗸𝘀, pushing the boundaries of computational capabilities ⚛️ - 𝗠𝗮𝗰𝗵𝗶𝗻𝗲 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗳𝗼𝗿 𝗮𝗻𝘁𝗶𝗺𝗶𝗰𝗿𝗼𝗯𝗶𝗮𝗹 𝗿𝗲𝘀𝗶𝘀𝘁𝗮𝗻𝗰𝗲 and 𝘀𝗼𝗰𝗶𝗼-𝗲𝗰𝗼𝗻𝗼𝗺𝗶𝗰 𝘃𝗮𝗿𝗶𝗮𝗯𝗹𝗲𝘀 𝗶𝗻 𝗲𝗽𝗶𝗱𝗲𝗺𝗶𝗰 𝗺𝗼𝗱𝗲𝗹𝘀, showcasing the power of data in healthcare and epidemiology 🏥 - Insights into the 𝗜𝘁𝗮𝗹𝗶𝗮𝗻 𝗲𝗹𝗲𝗰𝘁𝗿𝗶𝗰𝗶𝘁𝘆 𝗺𝗮𝗿𝗸𝗲𝘁 and the impact of data science on energy systems ⚡ Working alongside incredible professors, researchers, and industry professionals, I had the opportunity to hone skills in 𝗲𝘃𝗲𝗻𝘁 𝗼𝗿𝗴𝗮𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻, 𝗶𝗻𝘁𝗲𝗿𝗱𝗶𝘀𝗰𝗶𝗽𝗹𝗶𝗻𝗮𝗿𝘆 𝗰𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝗼𝗻, and 𝗰𝗼𝗺𝗺𝘂𝗻𝗶𝗰𝗮𝘁𝗶𝗼𝗻—all while delving into cutting-edge developments at the frontiers of science and technology. I’m grateful to everyone who participated and contributed their expertise, making this workshop a vibrant exchange of ideas. It was inspiring to see how 𝘱𝘩𝘺𝘴𝘪𝘤𝘴, 𝘥𝘢𝘵𝘢, 𝘢𝘯𝘥 𝘵𝘦𝘤𝘩𝘯𝘰𝘭𝘰𝘨𝘺 are being combined to address the challenges of our time. 💡 This experience has reinforced my belief in the transformative power of 𝘪𝘯𝘵𝘦𝘳𝘥𝘪𝘴𝘤𝘪𝘱𝘭𝘪𝘯𝘢𝘳𝘺 𝘢𝘱𝘱𝘳𝘰𝘢𝘤𝘩𝘦𝘴 and 𝘤𝘰𝘭𝘭𝘢𝘣𝘰𝘳𝘢𝘵𝘪𝘷𝘦 𝘱𝘳𝘰𝘣𝘭𝘦𝘮-𝘴𝘰𝘭𝘷𝘪𝘯𝘨 to shape a better future. If you’re interested in discussing more about the intersection of 𝗽𝗵𝘆𝘀𝗶𝗰𝘀, 𝗱𝗮𝘁𝗮 𝘀𝗰𝗶𝗲𝗻𝗰𝗲, 𝗮𝗻𝗱 𝘁𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝘆, feel free to connect or reach out! 🌍✨

Dear colleagues, As part of the 18th International Conference on Parallel Problem Solving From Nature (PPSN 2024), a prime international conference on evolutionary and other randomized search heuristics, we are organizing an informal workshop to celebrate the 30th birthday of estimation-of-distribution algorithms (EDAs), defined here as the 30th anniversary of the definition of the population-based incremental learning (PBIL) algorithm in 1994. We invite researchers and practitioners in the area of estimation-of-distribution algorithms to contribute a short talk to this workshop, covering current or recent own research, discussing important classic results and their impact on the field today, or proposing novel research directions. If you would like to give such a talk, please send us a short abstract by August 7, 2024. For details on the submission process and more information on the workshop, please consult the workshop page at  https://lnkd.in/gheVZh6T All the best Concha Bielza, Universidad Politécnica de Madrid Benjamin Doerr, École Polytechnique John McCall, Robert Gordon University Zheng Weijie, Harbin Institute of Technology, Shenzhen #PPSN #EDA #EvolutionaryComputation

📢 Workshop on AI, ML & Small Data Challenges in Statistics at GMDS Summer Academy in Munich!   We’re excited to announce that Moritz Hess, our Principal Investigator, will be joint-leading a workshop at the German Society for Medical Informatics, Biometry and Epidemiology (GMDS) Summer School in Munich from September 23rd to 26th. The workshop will focus on the practical application, interpretation, and challenges of using AI and ML in Statistics, particularly with small data sets. Register now to secure your spot! #AI #ML #Statistics #Workshop #GMDS #Munich #DataScience #SmallData

Project Discovery, an EVE Online initiative that has contributed to hundreds of years of medical research, is paving the way for the future of scientific innovation, according to Jérôme Waldispühl, Associate Professor of Computer Science at McGill University: “This community is really contributing to changing society by basically bringing this collective force to infuse science in society through a digital game.” Read on from Rachel Kaser on VentureBeat below: https://lnkd.in/gn49NtX8

I’m grateful to be part of the 18th Undergraduate Research Forum (UGRF) for the second time, collaborating with an amazing team on our project titled "AI-Generated Data for Enhanced Chronic Kidney Disease Classification with ML Models." ❤️ This research addresses the lack of medical data in chronic kidney disease by utilizing various machine learning models, including Random Forest, Decision Trees, Logistic Regression, SVM, and MLP, while evaluating performance through metrics like F1 score, precision, recall, and accuracy.📚 This experience has significantly expanded my knowledge and commitment to meaningful contributions in my field.❤️ #UGRF18th #MachineLearning #research #NileUniversity #ChronicKidneyDisease #medicalData

** 3 Days of Pure Innovation! ** Just wrapped the most intense and exhilarating hackathon of my life at #UDelATOMHackathon 🚀 (and the #ATOMHackathon)! Over the past whirlwind 3 days, I dove headfirst into the fascinating world of machine learning and computational biology. Here's why it was EPIC: Teamed up with brilliant minds 🧠 ! We had a phenomenal crew from different institutes, all working together to tackle drug-induced liver injury (DILI). Our goal? Building predictive models based on biochemical assay data – pushing the boundaries of tech and drug discovery! **Shoutout to my amazing mentor, Amanda Paulson, PhD! 🙎♀️ ** Her lightning-fast 5-minute machine learning for computational biology tutorial was a game-changer! ⚡ It totally demystified this complex area for me. Beyond grateful for this experience! Pushed my limits, expanded my skillset in a completely new domain, and connected with incredible professionals on the same innovation journey. Bonus highlight! I also had the chance to meet the incredible Eric Stahlberg 🤝 and learn about his insights on digital twin. His insights were truly amazing and opened my eyes to the future of personalized medicine! This is just the beginning! I can't wait to apply these insights and keep growing 😀 . #MomentInScience #ATOMScience #CancerResearch #MachineLearning #DrugDiscovery

Yesterday I had the pleasure to present with the Education and Training Foundation, at the T Level Health and Science Virtual Group, about "Regulation in Healthcare Software". If you haven't heard about the T-Levels, they are new two-year technical programmes. Their primary purpose is to prepare students for entry into skilled employment, or higher-level technical study. It is so important for the younger generations that there are different and exciting paths of education, rather than A-levels and university. #tlevels #educationforall

The perspective paper from our Immune Digital Twin Consortium is out! 🎉 Check it out here: https://lnkd.in/dFgEubBF What started as a 3-week workshop last spring has now become an amazing and motivated interdisciplinary community that works together to solve challenges related to #DigitalTwins. In our paper, we discuss the potential of digital twins, their basic principles but also the challenges and limitations that we have to tackle and be aware of. If you want to contribute with your expertise in any of the above and join us in this journey, please reach out and stay tuned for our future activities! 💻 🧬 🧪 🥼