NOETIK

Groundbreaking biology using NOETIK first-in-class foundation models of human cellular and tissue biology now live at #SITC24 and online. Come along for a quick tour: Lacey Padrón presents our full human discovery engine, from proprietary multimodal data generation (1,000 lung cancer patient samples) to the data pipelines and the AI discovery engine. This is one of the largest paired multimodal datasets of cancer biology! OCTO is a massively multimodal, transformer-based world model trained on data from all of our tumor samples. We use OCTO to discover T cell functional phenotypes in immune therapy resistant STK11 mutant tumors. Then, using simulated perturbations, virtual screen for therapeutic targets that are predicted to alter these T cell phenotypes to that of immune therapy sensitive tumors. In a second poster, Yu (Phoebe) G. shares exciting work studying Tertiary Lymphoid Structures (TLS) across our massive dataset of 1,000 lung cancers. TLS are immune structures that promote anti-tumor immunity. Can we discover therapeutic targets that enhance these structures in tumors? Because OCTO is trained on multimodal histology data, we can train a classifier on its embeddings to discover TLS across 10's of thousands of tumor images (Yes, it can also perform digital pathology tasks). TLS that OCTO finds confirm known biology (such as LTB and others). Then, using OCTO's simulation engine, we model maturation of these structures and discover potential targets to enhance their function. That's the summary, find our team at SITC or posters available here:

NOETIK is excited present results using our first-in-class foundation models of cell and tissue biology this Friday at #SITC24. Lacey Padrón (abstract #1231) will share a full stack view of our platform from data to discovery: https://lnkd.in/g2dXFthr Yu (Phoebe) G. (abstract #861) will share our work using multimodal foundation models to study the biology of Tertiary Lymphoid Structures (TLS): https://lnkd.in/g7c9uhyE For partnering/collaboration discussions, connect live with Shafique Virani, MD FRCS. https://lnkd.in/gpHCDpmn

Thrilled for NOETIK to be selected from among thousands of startups to participate in the 2024 Amazon Web Services (AWS) Generative AI Accelerator. We're training first-in-class foundation models of tumor biology on proprietary human data to solve the most critical problem in therapeutics: which drug, for which patient. Excited for our team to solve fundamental ML challenges to scale these models. https://lnkd.in/gFAwnv8M

Its hard to top such an exceptional piece by Andrew Dunn in Endpoints News yesterday on NOETIK $40 M Series A. But we wanted to share more about the rich dataset that this incredible team has been building for almost 24 months now. Check out our latest technical report: Tackling Cancer as a Data Problem, to learn more about our proprietary multi omics atlas of human cancer biology. Grateful for the hard work of the team across the company to build this. Industrial scale data generation begins in the wet lab, but requires a talented team across engineering, data science, computational biology to bring together. https://lnkd.in/ereAZVxS

Enjoyed speaking with Andrew Dunn to announce NOETIK $40M Series A led by Polaris Partners, Amy Schulman. The NOETIK team has made incredible progress in less than 2-years building a complex data platform and unique AI capabilities. We’re now ready to direct these capabilities toward therapeutics. Let’s go team!

Today we are releasing a first glimpse into the work we've been doing internally at NOETIK and sharing some details around OCTO, the first world model of cancer and the tumor microenvironment. OCTO is a massively multimodal, transformer-based model trained on proprietary data from thousands of patients’ tumors purpose-built for self supervised learning at NOETIK. It is truly multimodal, ingesting data that includes paired H&E, spatial protein, spatial transcriptomics, and whole exome sequencing for each patient. We've desigend OCTO to operate as a simulation engine to power diverse drug discovery tasks from target identification, to biomarker discovery and patient stratification, even prediction of therapeutics response populations. To find out more about OCTO, read our first?technical report: https://www.noetik.ai/octo

Excited to announce another incredible addition to NOETIK team! We are absolutely thrilled to welcome Carl Ebeling to lead our technical operations. Carl has a ton of experience in massive volume data generation, most recently as VP, Head of Production at Terray Therapeutics. Prior to Terray, he was Director of High Throughput Operations and New Process Implementation at Recursion. Carl has already started important work to increase the data generation capacity of our platform to feed more rich multimodal data to our world models. So excited to work with you again Carl!

Big news today: Incredibly thrilled to announce Shafique Virani, MD FRCS has joined NOETIK as Chief Business Officer! Shaf is a world-class leader in drug discovery with decades of experience at innovative large organizations such as Roche, and trailblazing biotechs such as Recursion & BridgeBio. While at Recursion, he led Corporate Development through transformative partnerships with Bayer and Roche/Genentech, including one of the largest and most innovative data-driven partnerships in the space. Shaf worked closely with Jacob and Myself, and was a key thought partner, in building the Oncology therapeutic area at Recursion, and again in recruiting critical leaders to establish the Neurosciences platform and therapeutic area. While he is an extraordinary commercial leader, he is also an experienced clinical development leader with a deep commitment to making an impact for patients. We're super excited to welcome Shaf to the team! Read more in our press release: https://lnkd.in/gHuFu3f9

We're looking for accomplished machine learning engineers to help scale self-supervised models on petabytes of multimodal biological data. If you love understanding distributed systems and have experience building infrastructure for large-scale training and inference, come work with us. Your work will fuel new discoveries in the biological sciences, with the potential to directly impact patients with cancer. Come build the future with us at NOETIK.

We are rounding out an incredible year at NOETIK. Still a couple days left until the holiday AND so much more going on still. But important to take a moment to reflect on LIGHTSPEED progress of this incredible team. This time last year, we did NOT even have a lab, now today we have built the full discovery platform we envisioned. Today, we've generated over 600 terabytes of human tumor data including spatial proteomics & transcriptomics, whole exome sequencing, H&E. At the same time we've launched perturb-map for scaled in vivo CRISPR screening. All these data are supported by robust data infrastructure built from the ground up. And these data are training a unique approach to multimodal vision transformers-based world models. And we have already seen that these models can do much more than image reconstruction, they can propose novel targets. Importantly, from our three ongoing projects, we are already starting to get to interesting target and therapeutics hypotheses. Congratulations to the team on a massively productive year! Excited to unleash this platform to keep building our immunotherapy portfolio next year! ??