Innovative TechBio Ideas (2nd part)
Innovative TechBio Ideas (2nd part)
AI/ML Revolutionizing Drug Discovery
For the first part of this newsletter:
Freshly caught news 🎣
🐟 Daewoong Pharmaceutical digested 800 million compounds to facilitate AI drug discovery
🐠 Exscientia: A Strategic Hold In The AI-Driven Drug Discovery Sector (Rating Downgrade)
🐡 This AI Startup (Receptor.AI) Joins Forces With NVIDIA's BioNeMo to Boost Drug Discovery
🪼 Iambic Therapeutics Unveils Advances in AI Platform to Streamline Drug Discovery
🦑 AI can speed up drug discovery but don’t expect it to cure cancer, yet
🐙 SYNTHIA SAS API: connecting retrosynthesis software with cheminformatics to expedite drug discovery
🦞 Ceremorphic's Pioneering AI Chips Redefine Future of Drug Discovery and Supercomputing
🦐 Artificial Intelligence / AI in Drug Discovery Research Report 2023 - Global Forecast to 2028: Market to Grow by $4 Billion, at a CAGR of 40.2%
🐬 AI Program Finds Thousands of Possible Psychedelics. Will They Lead to New Drugs?
🦀 FB1006: AI-discovered drug advances to clinical trials for ALS treatment
Alpenglow Biosciences
Alpenglow—founded as Lightspeed Microscopy, Inc in 2018 and rebranded in 2022—combines proprietary imaging technology, cloud computing and AI analysis to shed new light on tissue analysis with an AI-powered 3D approach. They use an open top light-sheet (OTLS) microscope with patented geometry that allows analysis of entire tissue samples or conventional multi-well plates. This fully automated system facilitates high throughput analysis, while cloud computing and AI-enabled software is allowing researchers to visualize and quantify complex spatial biology applications in 3D.
🧑🔬 In 2023, the Seattle-based Alpenglow Biosciences announced a collaboration with Mayo Clinic to adopt the firm’s 3D spatial biology platform in order to accelerate drug development and clinical diagnostics.
🧑🔬 In 2024, Alpenglow Biosciences announced the development of the 3D Derm Score Assay to reveal the complexities of the skin microenvironment and determine the relationship between critical structures in the epidermis and dermis. These structures are pivotal in driving skin diseases like Atopic Dermatitis, Hidradenitis Suppurativa, Alopecia Areata and Prurigo Nodularis.
Watch here how Alpenglow did a 3D projection of the scan of a whole skin punch biopsy stained with TO-PRO-3 (Nuclei), PGP9.5 (Nerves) and CD45 (Lymphocytes). By going deeper into the tissue, they were able to get the finer details, such as nerve structures and dimensions, and study the immune infiltration of lymphocytes and their distance from nerves.
The company was spun out of Washington University by Dr Nicholas Reder, Dr Jonathan Liu, Dr Adam Glaser and Dr Larry True, and so far has raised a total of $5.6M.
CleverPoint
CleverPoint in Poland creates hardware to understand brain processes using VR-integrated wearable technology. CleverPoint records the electrical activity of brain in the prefrontal cortex (EEG), the eye movement and the facial muscle movement (EMG) and the electrical activity of the heart (EKG) in order to analyze the user’s physiological response to VR content.
The real-time biosignals that are monitored include raw data (Amplitude/FFT, fast Fourier transform (FFT) is an algorithm that computes the discrete Fourier transform (DFT) of a sequence, or its inverse, IDFT), pulse, concentration (CleverPoint Algorithm), stress level (CleverPoint Algorithm) and labels from the VR-scenario. Then during offline data analysis the software is able to calculate:
level of alpha-, beta-, gamma-, theta brain waves,
indicators of heart rate variability,
level of concentration/relaxation, and
stress leveles.
The device costs €2800 for a kit composition of a CleverPoint device, a Facial Interface, a ECG cable and an Annual license on CleverLab.
By Leonard Wossnig CTO at LabGenius | Angel Investor | Honorary Research Fellow in Computer Science at UCL
“The first results of the Critical Assessment of Computational Hit-finding Experiments (CACHE) competition are out and have revealed that AI, while having a place in the drug discovery setting, is not yet the panacea for hit finding.
With the large number of new methods being published every week in the field of computational antibody engineering (specifically de novo design), it becomes hard to assess what works and what doesn't. Most papers sparsely cover the way the model details and evaluation process. Even fewer have experimental validation. Due to these reasons it becomes incredibly hard to assess what the practical impact on real-world drug discovery programs is.
I hence wanted to advocate for a similar effort to the CACHE competition in the area of de novo antibody or protein design. I am confident that everyone - companies, researchers, and investors alike - would benefit from it.“ From