Fernando José Suarez Saiz

University of Toronto simplifies protein interaction measurement technology Researchers at the University of Toronto have developed a platform called SIMPL2 that simplifies the study of protein-protein interactions, a crucial aspect of biological processes and disease. Led by Professor Igor Stagljar and Senior Research Associate Zhong Yao, the team designed SIMPL2 to optimize the measurement of these interactions for targeted drug therapies. This innovation addresses the challenge of protein-protein interactions being difficult to control with small molecules. The platform uses the split luciferase enzyme for detection through luminescence, allowing for more reliable measurements at a lower cost. Yao and Stagljar's work builds upon the original SIMPL system, improving its efficiency and sensitivity. The team plans to collaborate with Alán Aspuru-Guzik's lab at the University of Toronto and Insilico Medicine, a leader in generative AI drug discovery, to study interactions that play key roles in diseases like cancer, utilizing quantum computers and AI to develop new drug therapies. #quantum #quantumcomputing #technology https://lnkd.in/eASmrijP

1

I think the quote is important about "chipping away at problems bit by bit". Bioprinting was overhyped 10 years ago - but the hard work is being done by groups led by Tamer Mohamed, Sam Wadsworth and Simon Beyer at Aspect Biosystems . Tessella Biosciences Inc. is working hard on high performance biomaterial inks for 3D printing with the goal of transitioning our R&D grade products to GMP for human use.

9

AI isn’t magic—it’s a tool; and it's transforming biotech right here in Toronto. Thrilled to share that I was recently featured in a Toronto Star article where I discussed how Intrepid Labs is leveraging the power of AI to radically reshape drug development. With a 90% failure rate in the industry, it’s clear that something needs to change—and we believe AI holds the key. At Intrepid Labs, we’ve developed a self-driving lab, where AI and robotics take on the heavy lifting, enabling us to rapidly identify the most effective drug formulations. 🗨️ As I mentioned in the article: "AI is not magic; it’s a tool." And when harnessed correctly, it can transform the future of biotech, not just for startups and multinationals, but, most importantly, for patients who stand to benefit from faster, safer and more effective treatments. I’m also looking forward to speaking at this year’s Elevate Festival where I’ll discuss the role of AI in drug formulation and why Toronto is the place to be for innovation in AI. 🚀 🔗 https://lnkd.in/g6ZYQxHx #AI #DrugFormulation #Innovation #Healthcare #DrugDevelopment #Toronto

749

8 Comments

I’ll be at NeurIPS in Vancouver this week presenting the findings of our latest paper at the ‘AI for New Drug Modalities’ workshop on Sunday Dec 15th. Immunotherapy development currently requires a trade off between speed, cost, and quality. While the application of machine learning has advanced areas of drug discovery, its impact in immunotherapy development remains limited due to the lack of standardised dataset. At Coding Bio, we’re pioneering a unique approach that combines high-throughput experimental techniques with ML to accelerate the future of cancer treatment. Our latest paper published on arXiv last week, demonstrates how protein language models (PLMs) – when fine-tuned using a preference optimisation task – show strong correlations with biological assays. These insights open up new possibilities for hit maturation in cell therapies like CAR-T, offering a faster and more effective route to better therapies for diseases like cancer and autoimmune disorders. You can read the paper "Harnessing Preference Optimisation in Protein LMs for Hit Maturation in Cell Therapy" here: https://lnkd.in/gUpM-6MZ Big thanks to all the authors Katarzyna (Kasia) Janocha, Annabel Ling, Alice Godson, Yulia Lampi, PhD, and Simon Bornschein! Reach out if you want to meet at NeurIPS! #CodingBio #AIxBio

117

2 Comments

Hold my beer, eh! This breakthrough was years in the making … and it’s big. For my non-scientist contacts I will try to explain simply why these Canadian researchers may have just changed the game for one important slice of biotech and pharma. It could mean a radical short-cut to better therapies and diagnostic tests, based on antibodies. Antibodies are naturally produced in the body in response to infection and other diseases. The immune system has evolved over Billions of years to do this effectively. In a span of just a couple of weeks it produces and tests perhaps 1,000,000,000,000,000,000 antibodies. If you want to visualize that number, imagine counting all the grains of sand on all the beaches and deserts on earth. The best ones are selected and produced to fight the disease. They are incredibly effective in moving through the body and binding to their target, then either simply blocking the offensive or recruiting other parts of the immune system. Antibodies can be used as amazing tools for healthcare. An antibody can be a made into drug that only targets specific diseased cells, or blocks a specific virus. Think of a test strip that shows a clear signal only when the antibody binds its target. Many of the biggest blockbuster drugs are antibody based. It’s no wonder then, that billions of dollars are spent trying to find the perfect antibodies to use as therapies for the next disease target. But how? We can try to replicate nature - generate sequences (using AI maybe), produce and test them. Or we can try to see if lab animals will produce useful antibodies - but then what? Will these synthetic or animal antibodies be safe to use in humans? Will they be stable and effective? Years of costly R&D is involved in resolving these questions and the failure rate is very high. For every successful new therapy there are hundreds of failed programs and billions of dollars spent. So why not start with antibodies that are already known to be safe, stable and effective in a human - simply by using antibodies extracted from a human patient who has recovered from the disease?! If you think this idea is almost too obvious, you aren’t alone. In fact, physicians have been using blood from recovered patients to treat sick ones for a long time (read about Convalescent Plasma Therapy). But there’s no scalable supply of such plasma. So the problem is finding the antibody you need out of incredibly complex human biology, so you can test it, produce more of it and distribute it to thousands of patients. The solution is to isolate and sequence the important antibodies, then study them and produce more of them. My colleagues have created groundbreaking technology for sequencing antibody proteins out of the complex mixture in human blood. In this Nature Communications paper they show how it works, and that antibodies they produced this way work well to neutralize COVID for example. #biotech #canadianinnovation #antibodies #drugdiscovery #nextgenmedicine

11

Two AI giants Alán Aspuru-Guzik and Alex Zhavoronkov meet in one room! Alex Zhavoronkov is currently presenting at the Acceleration Consortium seminar at the University of Toronto. Don’t miss this opportunity to hear about the latest advancements in #drug development using #AI technologies at Insilico Medicine! Happening now: https://lnkd.in/gU6F_5fG #drug #discovery #AI #ML #aging #acceleration #science #insilico

52

Exciting Project Completed!🚀✨ I recently had the opportunity to lead a critical upgrade and migration project at BC Cancer's Genome Sciences Center, which involved moving our internal chat platform, Rocket.Chat (RC), from a legacy CentOS 6 environment to a new Rocky Linux 9 setup. RC is an open-source chat tool that we rely on for our internal communications. However, it had become outdated, and the legacy version was no longer supported by the latest RC mobile application and Linux environment, limiting its functionality and security. As part of this project, I successfully built and configured new servers running Rocky Linux 9, deployed the latest RC version for full compatibility, and migrated all our old chat data to the new servers. Here’s a snapshot of how I achieved it: 1. Thoroughly studied the vendor documentation to fully understand the build and migration process, ensuring the correct MongoDB and Node.js versions were implemented in the server configuration, for compatibility with the latest RC version. 2. Built and configured three Rocky Linux 9 servers on VMware vCenter—two dedicated for hosting RC and one for the database. 3. Compiled RC from source using Node.js and deployed four instances across two servers to address the single-threaded limitations of Node.js. 4. Migrating the legacy chat database was the most challenging task. It required moving up different versions of MongoDB and RC iteratively, each set up on a different dev server, while also making sure that there was no data loss. 5. Configured Nginx as a reverse proxy on the servers to efficiently distribute traffic across multiple RC instances. 6. Implemented SSL certificates using Puppet for secure and encrypted communication. 7. Automated tasks like server restarts and backups with shell scripts, streamlining maintenance processes. 8. Led extensive testing phases, gathering user feedback on JIRA, and implementing improvements based on this feedback. Also created detailed user instructions for the updated platform. 9. Integrated RC with Bamboo via webhooks, optimizing DevOps workflows by enabling real-time notifications. 10. Decommissioned the legacy VMs after successful migration to the new environment. Key outcomes: 1. Successful transition with minimal downtime. 2. Enhanced system performance, security, and feature set. 3. Automation of routine tasks. Learnings: 1. Importance of User Feedback: User input was key to improving the experience and ensuring smooth adoption. 2. Adaptability: I developed the ability to quickly adapt to unexpected challenges during the migration process. 3. Attention to Detail: Ensuring data integrity highlighted the importance of precision in every step. Conclusion: This project not only enhanced my technical skills but also helped me understand the importance of extreme ownership, fostering my ability to work under pressure, learning on the go, and being resilient. I'm eager for the next challenges and to continue learning!

28

6 Comments

Join us at #JPMweek to discover how we leverage AMICA™, our multi-omic AI model of the immune system, to enhance data-driven drug development decisions. #AI #immunology

84

🔬 Exploring the Impact of #PIK3CA Mutations on Breast Cancer Treatment 🔬 Our recent study deals with the complexities of the PI3K signaling pathway in breast cancer (BC), focusing on how PIK3CA mutations influence therapy resistance, particularly in #HER2-positive BC. We analyzed 1691 early BC patients across four #neoadjuvant multicenter trials: #GeparQuattro, #GeparQuinto, #GeparSixto, and #GeparSepto. Of these patients, 17.9% had PIK3CA mutations, with prevalence varying across subgroups: 23.5% in luminal/HER2-negative, 20.8% in HER2-positive, and 7.9% in triple-negative BC. Our findings revealed that mutations in exon 20 are linked to a poorer response to anti-HER2 treatment, especially in HR-positive HER2-positive BC. In contrast, exon 9 hotspot mutations (p.E452K, p.E545K) did not significantly impact therapy response. Interestingly, patients with residual disease after neoadjuvant treatment had better #survival rates if PIK3CA was mutated. In conclusion, the exon 20 mutation (p.H1047R) is associated with lower pathological complete response rates in HER2-positive BC, while exon 9 mutations seem to have less impact. 📰 Find here our recent publication: https://lnkd.in/eGDPmwQW   We are conducting our #GeparPiPPa (GBG 105) study to investigate precisely this group of patients with HR-positive, HER2-positive, PIK3CA-mutated BC. 🎗 This collaborative study between GBG Forschungs GmbH, Stiftung "International Breast Cancer Study Group IBCSG" and @AGO-B study group is a randomized, multicenter, prospective, open-label phase II trial comparing neoadjuvant endocrine therapy in combination with trastuzumab, pertuzumab +/- the PI3K inhibitor inavolisib. 🚀 Recruitment has already started. If you are interested, please have a look at our website: https://lnkd.in/ewk4UaZw   #BreastCancerResearch #PIK3CAMutations #Oncology #HER2Positive #NeoadjuvantTherapy #CancerGenomics #MedicalResearch #Inavolisib #ClinicalTrials

24

How can I explore some ideas for novel therapies? Here is my dilemma. I was in the biotech industry for 20 years, about 15 at senior levels (Director of Research, VP Preclinical). Then working for IRAP in the Canadian Federal Gov for 9 years full-time, then called back from 2020 to 2022 part-time as an technical and business assessor for many Federal and Provincial Programs. I have some ideas regarding novel therapies and other projects but I do not have access to lab facilities. Without those it is very difficult to do crowdfunding; without incubator facilities likewise. Also, without data, I don't believe that Angels or VC's here in Canada would provide funding for an idea only. I am looking for ideas or collaborators. Does anyone have ideas? Anybody interested in setting up a commercial incubator in Vancouver? Any other suggestions? Please message through Linkedin and all messages greatly appreciated. Michael

1

AI Proteins, a biotechnology company that uses computational de novo protein design to create therapeutic miniproteins, announced last week an agreement with Bristol Myers Squibb (BMS) to discover and develop novel miniprotein-based therapeutics using AI Proteins’ discovery platform. The deal could be worth up to $400M plus royalties on net sales. Miniproteins comprise: a peptide that folds into a highly stable structure, optional disulfide bonds, canonical amino acids without post translational modifications, entirely novel composition of matter designed to be non-immunogenic, and more. Notably, the company says, “because drug-like properties are designed into the miniproteins from the start, every molecule with optimized activity is ready to begin pre-clinical studies.” AI in drug discovery is one of the most active areas of biotech deal-making right now. There has been a steady stream of deals the last two years, but just within the last few months:

2

Loved talking about how to best train people for the bioeconomy. Thanks for the opportunity Ontario Genomics. A few things I spoke about and also learned: - Project and financial management skills are crucial. 💲📊 - In the design-build-test-learn cycle, skills in "test" are the most important. A lot of "build" will be, or already is, automated 🤖 - Experiential projects and team-based learning are key. Find every opportunity to try this out. No "real world" project is done alone. 👍 - Advocate for yourself! 🙂

45

2 Comments

It's a real pleasure to share this announcement for the first cohort of the Canadian Neuroanalytics Scholars Program! Together with partners at Campus Alberta Neuroscience, The Neuro (Montreal Neurological Institute-Hospital), and Ontario Brain Institute this is a next generation opportunity for forward-thinking research solving real challenges in brain health using multi-modal data and advanced analytics. We can't wait to see what comes out of these great projects and talented scholars! #neuroscience #analytics #brain #ai #datascience #data

38

data cleaning and wrangling (the mundane work) are what you need to be good at. My odyssey of obtaining scRNAseq metadata https://lnkd.in/eB-kDGeD

20

1 Comment

This article is inspired by comments from Tony Czarnik and Stefan Ivanov regarding my previous blog post at https://lnkd.in/gFPhXfKW. To clarify, I am writing from the perspective of an AI believer or practitioner, acknowledging the critical viewpoint that claims "AI has shown little in terms of actual achievements in drug discovery." However, I believe it is not sufficient to simply agree with this stance without further elaboration. My focus here is on higher-level concepts rather than practical examples. Here are three key takeaways: 1. Need for Comprehensive Data Collection The current drug discovery process often lacks systematic documentation and relies heavily on subjective reporting, which hampers the ability to draw robust conclusions. Scientists, particularly chemists and biologists, should prioritize automating the collection of extensive data and metadata to create a rich database for analysis. 2. Integration of Mathematical and Data-Driven Approaches Two primary strategies for improving drug discovery are mathematical modeling and data-driven methods. While traditional scientific practices lean toward mathematical modeling for deeper understanding, advancements in AI highlight the effectiveness of data-driven approaches in addressing existing challenges and providing new insights into complex biological processes. 3. Transformative Role of AI AI technology plays a crucial role in drug discovery by enabling scientists to embrace data-driven methodologies that enhance problem-solving capabilities and deepen understanding of biological systems. Combining both mathematical models and data-driven techniques can lead to a more robust framework for innovation in the pharmaceutical industry.

17

21 Comments

Check out our paper how we are revolutionizing data sharing using GenAI to understand informed consent documents. This work could easily be widely leveraged across the industry and with other data use agreements allowing appropriate data reuse. #TeamBiogen

14

Curious how to accelerate drug development with our platform ? Listen in for some exciting results at #ASMS2024, Thursday afternoon session: Drug Discovery and Development: Qualitative and Quantitative Analysis #drugdevelopment #quantitativeanalysis #offtargets

12

A Vancouver biotech working on bioprinted tissue therapeutics has reeled in $115 million via a Series B. The capital will support Aspect Biosystems' in-house programs in endocrine and metabolic disorders while also working on diabetes with partner Novo Nordisk. Nan Li, a managing partner at the round's lead investor Dimension and new Aspect board member, called it a “modality that’s too elegant to ignore.” “It’s one of these modalities that the industry continues to invest in because of its promise. We’re kind of chipping away at the problems bit by bit but even with some recent clinical challenges in regenerative medicine, a lot of large pharma companies continue to make bets there and put it on their 10-year roadmap for where they see the industry going,” Li told Endpoints News More below from my interviews with Aspect CEO Tamer Mohamed and Li: #aspectbiosystems #novonordisk #tissuetherapeutics #bioprintedtissue #metabolicdisorders #endocrinedisorders #biotechvc #biotechfunding #canadianbiotech

144

4 Comments