Innovative TechBio Ideas
Innovative TechBio Ideas
AI/ML Revolutionizing Drug Discovery
Welcome back to another edition of MetaphysicalCells 🐚🪸
Freshly caught news 🎣
🐟 Google DeepMind alumni unveil Bioptimus: Aiming to build first universal biology AI model
🐠 Bioptimus nabs $35M to accelerate biology research with AI
🐡 Avance Clinical adopts Ryght AI tech to streamline clinical trial processes
🪼 Serbia to Invest €76.8 Million in AI Biotech Master’s Programmes
🦑 Healthcare startup Abridge raises $150 mln for AI model for clinicians
🐙 Akin IP Lawyers Publish AI-Assisted Drug Discovery Article in PharmaPhorum
PredxBio
PredxBio (previously known as SpIntell), founded in 2017 by Michael Becich, S. Chakra Chennubhotla, D. Lansing Taylor and Jeffrey Fine in US, is leveraging
unbiased spatial analytics—namely pathology, spatial transcriptomics and spatial proteomics to identify cell states & fusion cell types, microdomains associated with disease progression & outcomes and pathway interactions & networks—and explainable AI on multi-hyper-plexed imaging data to get to the “why” in immunotherapy trials.
Their PredxBio Platform—a powerful analytical software as a service, SaaS, application—is offering:
🔶 PredxNeuro,
🔶 PredxOnco and
🔶 PredxClinical analysis,
by using cell segmentation & basic quantitation, spatial cell phenotyping, microdomain discovery from multi-hyperplexed & multi-modal image datasets and microdomain guided network biology. In this way, they can predict tumor initiation, progression and response to immunotherapy thus enabling better outcomes for patients, by identifying sub-populations of patient response to immunotherapy with >90% confidence. Their solutions empower clinicians/researchers to accelerate drug discovery, optimize clinical trials, advance companion diagnostics, personalize therapeutic options and improve clinical workflows.
In 2019, they received the HistoMapr SBIR Award and an investment from America's Seed Fund. In 2020, they established a partnership with CellNetix Pathology & Laboratories. In 2021, they established a partnership with Allegheny Health Network and they received an investment from Innovation Works. Moreover, they have 20+ peer-reviewed publications and a dominant IP position for adaptive and agnostic solutions (for transmitted light and multi-hyperplexed and any imaging platform). Plus they have a collaboration with the leading academic medical center Roswell Park Comprehensive Cancer Center. PredxBio raised a total of $3.1M and the CEO is Dusty Majumdar, PhD.
AdaptyvBio
Adaptyv Bio in Lausanne, Switzerland is building a next-gen protein full-stack platform to allow you to synthesize and test any protein you design (based on novel algorithms such as Alphafold or RFDiffusion), integrating AI at every step of the protein engineering cycle: from designing sequences, to planning experimental protocols, to running those experiments autonomously (with microfluidic chips) and finally tracking all data and metadata. Then all measurement data is automatically processed and uploaded to allow users to improve their ML models with each experiment.
To do so they have launched Automancer, a software application which makes it possible to automate lab protocols. The protocols designed with Automancer typically interact with lab devices, such as temperature controllers, relays, microscopes and liquid handlers. Automancer, which is open-source, is highly extensible and can also support custom actions by creating your own plugins and developing your own connections for devices.
For example, a protocol designed with Automancer is about flow reagents on a microfluidic chip, controlling both external rotary valves and on-chip microfluidic valves. As a matter of fact, at the heart of Adaptyv Bio‘s protein platform are customized setups that miniaturize all processes (that usually require multiple laboratory machines in a real lab) by using tiny microfluidic chips that consume 1,000 times fewer reagents than any available alternative. In order to generate a stable microfluidic droplet stream it this setup—by controlling the pressure of the droplet phase vs the pressure of the encapsulation phase—Automancer can easily write a function that decreases the droplet phase pressure periodically and then use a microscope camera to record images of the droplet stream, identifying individual droplets and measure their size.
Adaptyv Bio, it’s not just an AI solution helping you to limit human intervention and error, improve reproducibility (creating closed loops during drug discovery) and run experiments out of office hours. But it is also a green AI solution, environmentally 💚 friendly, reducing laboratory waste.
The development of Automancer was originally born out of a collaboration with the Maerkl Lab at EPFL and Adaptyv Bio founded in 2021 is now backed by EPFL, YCombinator and Wingman Ventures and has raised a total of $3.1M. Its founders are Julian Englert (CEO), Moustafa Houmani (CTO), Daniel Nakhaee-Zadeh and Amir Shahein (CSO). Amir, the co-founder and CSO of Adaptyv, is also the Head of Bio in Julian Capital and co-founder in a stealth startup.
Julian Capital (partners Mat Vogels, Julian Shapiro) is a seed VC who writes $500K checks focused on deeptech, and invests in:
Elicit: that analyzes research papers at superhuman speed. And
Parallel Bio: that finds the fastest path to drugs that work for more people by starting in human systems first. Their platform replicates the human immune system in a dish across populations to discover cures at a pace never seen before.
Xbiome
Xbiome, founded in 2017, is the leading AI-based gut microbiome drug development company in China, driven by AI and data while building an AI+BT microbiome product development platform to achieve high-quality data analysis, and output for high-quality microbial drug pipelines and products.
Xbiome has established bioinformatics teams in China and the United States, with a global multi-center layout centered in Shenzhen and Boston. After years of development, five core platforms have been developed by Xbiome:
🔶 Bioinformatics Platform, for AI drug discovery and Clinical Analysis.
🔶 AI Microbial platform, an efficient microbial culture platform including an automated and selective bacteriological technology platform, a bacterial physiological and biochemical identification platform, an anaerobic microbial culture and fermentation platform, and an intelligent strain library information management system.
🔶 Molecular Platform, currently, it has a complete intestinal microbial detection and analysis platform based on next-generation sequencing, qPCR and other methods. And it has also developed a multi-tissue (tumor, saliva, etc.) bacterial flora detection technology.
🔶 Preparation Platform: for oral freeze-dried microecological preparation, oral novel controlled release dry suspension preparation and oral enteric-coated microsphere preparation. And
🔶 Production Platform: under the guarantee of an excellent production platform and quality system, in June 2021, an FMT drug from Shishijun obtained IND approval from the US FDA, marking the company's official entry into the clinical stage.
In 2024 Xbiome:
won the top award 🏆 among the top 50 biotech innovation companies in the Greater Bay Area,
the research results of Xbiome and Beizhuang were published in the Cell sub-journal, and
Xbiome joined forces with Qianhai Shekou Free Trade Zone Hospital to enable intestinal bacteria transplantation to benefit patients.
Xbiome raised $124M so far and the founder is Yan Tan.
📢 “It is a pleasure to see AUP1602-C become Xbiome's first project to enter Phase II clinical trials, marking a new stage for Xbiome.” Jan 22, 2022, Yan Tan
AUP-16 (AUP1602-C) is a multi-therapy live bacterial vector for chronic wounds: Diabetic Foot, Leg & Pressure Ulcers, by Aurealis Therapeutics, a partner of Xbiome.
CarbonSilicon
CarbonSilicon AI founded in 2021 in China, is a technology company focusing on AI drug development. Carbon Silicon Intelligence is providing customers with open and easy-to-use training frameworks, models, data and automated robots, and is empowering partners to develop independent and controllable intelligent solutions.
The startup provides
Inno-Docking, a complete protein preparation, ligand preparation and intelligent setting of docking parameters.
Inno-Rescoring, featuring AI-scoring functions to evaluate protein-ligand binding affinity.
Inno-ADMET for ADMET properties, ChemFH to filter frequent hit compounds, and Inno-SA to predict substructure-related toxicity. And many more.
Carbon Silicon was founded by Deng Yafeng, Hou Tingjun, Shi Hui and Cao Dongsheng.
Meet the World’s #1 AI-Powered Presentation Generator 👇
Data2Discovery
Data2Discovery (Utah US, 2012) uses AI to find hidden connections and new insights in diverse, linked datasets by allowing researchers to understand and treat diseases by connecting data in new ways.
In 2021, Data2Discovery along with Indiana University Crisis Technologies Innovation Lab and two partner companies Disaster Tech and OPS have been awarded a $2.3M contract from the US Army Telemedicine and Advanced Technology Research Center to create a Technology in Disaster Environments Learning Accelarator.
In 2022, Data2Discovery discovered a promising potential treatment for mycobacterium infections including Mycobacterium tuberculosis (MTB) and Mycobacterium Abscessus Complex (MABC). The treatment shows special promise for MABC, an increasingly impactful disease for which there is no current effective treatment.
Aivia
Aivia (acquired by Leica Microsystems in 2021 from Svision LLC) combines powerful AI-guided image analysis and visualization solutions for data-driven scientific discovery. Using state-of-the-art algorithm and software architecture, Aivia delivers top performance on critical tasks such as display of large images and analysis of complex biological phenomena, and is powered by a range of ML technology for both image segmentation, object classification and novelty detection.
Aivia offers:
✨ A complete DL workflow, from ground truth generation to model training and analysis, with Aivia 8 and Aivia Cloud.
✨ Aivia 8.8, to further enhance the ML-enabled Pixel Classifier. By painting a few sections (in 2D or 3D) of an image, Aivia enhances and segments objects automatically.
✨ A library of pre-trained deep learning models to apply them to your data.
✨ Aivia 9.5, to integrate DL into your image analysis workflow on any recipe, opening up a whole new world of possibilities to process your images.
✨ Aivia 10, providing AI access for all with the launch of Aivia That Learns, that predicts suitable image analysis parameters for 3D Object Analysis and 3D Neuron Analysis recipes for your data. This enables a reduction of 74% in analysis time by streamlining the process to optimizing the analysis parameters.
✨ Aivia introduced a revolutionary new way to display super-large datasets and interact with hundreds of thousands of 3D objects in real-time, by implementing an innovative multi-resolution, multi-block file architecture for volume rendering; and an efficient memory management system with optimized GPU usage for surface rendering. In Aivia 8.5, volume rendering is further refined - allowing you to display large teravoxel datasets in seconds and interact with individual objects in real-time without lag.
✨ Aivia is also pioneering the latest virtual reality (VR) technology to bring you immersive image exploration. You can view and interact with your 3D and 4D datasets in the VR space to examine and compare objects.
✨ Aivia currently features 17 application-specific image analysis pipelines called "recipes" for 2 to 5D microscopy images. The recipes incorporate the latest high performance algorithms for image enhancement, segmentation and tracking (where applicable) thus enabling you to accurately detect all objects of interest on your images.
✨ Aivia 12, incorporates Aivia Elevate CellBio that includes specialized AI tools for analyzing complex interactions in cellular environments such as organelle organization in a cell (AI-enabled 3D tracing of Golgi-stained images) or the number of cells within an organoid. And
✨ Aivia Elevate Neuro, provides tailored AI tools for image visualization and analysis to automate challenging neuroimaging tasks, such as neuron tracing, dendrite segmentation and more.
📢 How tech will shape biotech in 2024
By Vega Shah, PhD 👸, Senior Product Marketing Manager, Benchling, and by Janet Matsen, PhD, Software Engineer, Machine Learning, Benchling
Ariadne.AI
Founded in 2016 in Switzerland, Ariadne is dedicated to automated biomedical image analysis, developing tailored image analysis solutions for screening assays. They offer:
🤹 Spatial, a cloud-based spatial omics in your web browser, with
no coding and no specialized hardware, that supports proteomics, transcriptomics, metabolomics and more.🤹 3dEMtrace: Image segmentation for electron microscopy datasets,
a new platform for segmenting cell structures in volume electron microscopy datasets by simply uploading your image stack and specifying the structures you would like to segment.
🤹 LMtrace: Neuron reconstruction made easy, a worry-free pipeline for reconstruction of neurons and glial cells in light microscopy datasets by just uploading your dataset.
🤹 Socomap, a fully automated whole-brain segmentation platform,
an online platform for analyzing whole-brain light microscopy stacks of mice. They provide highly accurate soma segmentation and cell counts for all cells expressing a specific marker. They also offer brain region specific cell counts by mapping the dataset to the Allen Brain Atlas. And
🤹 Tomotrace, a high-throughput tomogram segmentation pipeline, a platform for automated segmentation of stained and unstained TEM tomograms. The pipeline delivers high quality segmentation for a large range of different organelles and other cellular structures, from microtubules to ER, Golgi apparatus and plasma membranes.
Founded by Fabian Svara, Jörgen Kornfeld and Adrian Wanner Ariadne has raised $2.4M.
🗣️ On February 29, 2024, join them for a market launch webinar.
📢 Trisonic Discovery is introducing the A1 Acoustic Dispenser. A revolution in acoustic dispensing technology. Their mission is to make non-contact, acoustic, low volume dispensing, affordable, reliable and in a compact bench-top form factor that is suitable to even the smallest lab.
A1 is an acoustic dispenser which permits plate to plate fluid transfers in nanoliter range and above.
As of Q1-2024, A1 is at an advanced prototype stage and they welcome interested parties who would like to beta test A1 to get in contact with them.
Trisonic Discovery is also looking to engage with visionary investors.
Enable Medicine
Founded in 2019, Enable Medicine, accelerates insight discovery by generating, managing, and analyzing spatial biology data on the Enable Cloud Platform. They offer,
🔶 The Enable Lab: measuring 51+ protein and RNA biomarkers on human and murine tissues to assess spatial relationships, providing a detailed window into disease biology and therapeutic action.
🔶 The Enable Atlas: integrating across genomics, transcriptomics, proteomics, imaging (Generalized Immunostaining, Hematoxylin and Eosin, Phenoimager, Imaging Mass Cytometry, Phenocycler-Fusion, Phenocycler) and clinical data to link historically disparate data types, leading to the discovery of novel biology.
🔶 The Enable Biological Operating System (BIOS): a custom, fit-for-purpose image processing and analytics suite that utilizes the latest advances in computer vision to understand cellular frequencies, interactions and neighborhoods at single-cell resolution.
In June 1, 2023, Enable Medicine announced the launch of Generative Biological Search (GBS), marking the first integration of large language models, biological data, and research workflows in the field.
Enable Medicine raised $60M in its latest funding round, which was Series A round held on Apr 07, 2022, for a total of $63,3M. The founders are Aaron Mayer and Sunil Bodapati, and the CEO is Kamni Vijay, PhD.
For more: Innovative TechBio Ideas (2nd part).
Until next time 🐞
Awesome content. Did you see this by the way?
https://26206544.fs1.hubspotusercontent-eu1.net/hubfs/26206544/BioStrand/Content/EpitopeBinning_CS_A6.pdf
Welcome.
I look forward to your feedback regarding this, looks fantastic!