Turbine

🎙️ Join our webinar co-hosted with Accenture: New technologies to accelerate ADC design! Antibody-drug conjugates (ADCs) are transforming cancer therapy, with 15 approved drugs on the market — but over 100 discontinued programs highlight the complexity of ADC design. How can we overcome these challenges to accelerate development? 🗓️ When: December 12, 2024 - 3pm (GMT) 🎙️ Speakers: Eleni Tokali (Host) Daniel Veres MD PhD (Co-founder & CSO, Turbine) Hassan Naseri (Scientific & Next Gen. Computing Lead, Accenture) 💡 Topics: 1️⃣ Overview of the ADC market and challenges of ADC design 2️⃣ How can simulations help in understanding payload response and matching them to the right patients? 3️⃣ How can AI-driven tools help in designing cleavable linkers for ADCs? Don’t miss this opportunity to explore the future of ADCs! https://lnkd.in/d_ifg_Kb

Honored to share that Turbine has been selected as a finalist for MSD's Innovation Award of the Citeline Scrip Awards 2024. Congratulations to Generate:Biomedicines for winning the category! Being recognized alongside such groundbreaking innovations is truly inspiring. This acknowledgment underscores the transformative potential of our Simulated Cell™ platform, and we’re proud to contribute to advancing new frontiers in R&D. https://lnkd.in/eaAKmrWF

We are thrilled and honored to be among the nominees for the #FutureUnicornAward! Huge thanks to the organizers and the incredible network behind this initiative. Let’s shape the future of European innovation together! 🚀✨

🧠 Why does a tech-bio team choose to cater a wide market instead of point solutions? 🧬 What impact does an integrated platform like this offer to biopharma companies? 🇭🇺 How do you build a global AI for bio success story from Hungary? Tune in for Philip Hemme of Flot.bio and our CEO Szabolcs Nagy discussing even more crunchy details with some sneak peek into their personal history. https://buff.ly/48MKakn

Current single-cell foundational models, including scGPT, can underperform compared to simpler approaches in predicting cellular responses to perturbations, highlighting significant limitations in existing benchmark datasets: 🌲 Turbine's benchmarking showed scGPT lagging behind simpler approaches like averaging training samples and Random Forest. 🐾 Models with biologically relevant features can significantly outperform scGPT. 🔻 Perturb-Seq datasets used for benchmarking are limited by low perturbation counts and lack of intra-perturbation variance. ⛓ Pseudo-bulked expression profiles outperform foundational models, suggesting little-to-no advantage in single-cell level modeling, especially in low-heterogeneity cell lines. Findings suggest revisiting benchmarking practices for more effective evaluation of post-perturbation gene expression prediction. Read more in the following preprint: https://lnkd.in/dKBGpYxr

"Hungary & #CentralEasternEurope is getting pretty good at nurturing tech startups. But any #DeepTech founder grappling with commercializing scientific breakthroughs faces too many barriers." Click to read more from our CEO Szabolcs Nagy!

Our dearest Anna Török-Böröcz has just secured her place among the top 🇭🇺 HR leaders of year 2024! 🎊

As elegant as ADC-based therapies promise to be for oncology, as hard it is to deliver them. Our obstacles range from the very depth of biological understanding to drug design. Turbine's simulations help in: 🧬 Identifying drivers of response for payloads to guide rational patient selection 🧪 Finding the right payloads for disease-subtypes to address ADC-resistance in the clinic Our team is visiting World ADC Summit in San Diego, presenting a poster during today's scientific session. Please see our poster by clicking the link below and feel free to reach out through [email protected]! https://buff.ly/3UFoyR3

It's not the first time we talk about the biases in training data that lead AI disease models astray. In this podcast episode we discuss the methodology we use to accurately assess computational model performance. We also share insights from our big pharma collaboration on how we use public, internal, and partner datasets to create patient avatars for simulations. Tune in to listen or visit benchmark.turbine.ai to learn more about the EFFECT Benchmark Suite! Let's work on finding ways to make more sense of ground truth data for better models of disease biology!

We believe that RNA sequencing (RNAseq) is the best data source currently to computationally model cell behavior. Our CTO, Kristóf Szalay explores this topic in his latest blog post. 💡 Key takeaways: 1️⃣ RNAseq is cost-effective and scalable, making it ideal for large datasets with multiple perturbations. 2️⃣ Baseline RNAseq data correlates well with proteomics. 3️⃣ Basic machine learning models achieve comparable performance when predicting post-perturbation cell viability using either RNAseq or proteomics features. 4️⃣ With the right approach, downstream gene expression reveals crucial insights about upstream signaling pathways. Dive into the details by clicking the link below!