BioByte 075: Next generation antibodies, CRISPR beyond the liver, 85 Million cells at your fingertips
Welcome to Decoding Bio, a writing collective focused on the latest scientific advancements, news, and people building at the intersection of tech x bio. Happy decoding!
It is now May. What have you achieved this year?
…now feeling panicked?
Don’t worry, just keeping reading to achieve some scientific enlightenment.
What we read
Blogs
85 Million Cells—and Counting—at your fingertips [Jeffrey Perkel, Nature News Feature, April 2024]
Biologists who work with single-cell gene expression data have faced a major headache—thousands of datasets scattered across the internet, processed differently, and using inconsistent cell nomenclature. The Chan Zuckerberg Initiative's new CZ CELLxGENE platform aims to change that. Launched earlier this month, it provides a centralized resource containing over 85 million uniformly processed single cells from 1,317 datasets across 844 cell types. Researchers can seamlessly explore this vast data repository through an online portal or programmatically via R and Python. The integrated data opens the door for unprecedented cross-dataset analyses and meta-analyses. Beyond data aggregation, the platform enables building sophisticated AI models by providing pre-trained versions that can be refined and applied to researchers' data. These models allow mapping new datasets into a common space to identify similar cells or conditions. And the capabilities are rapidly expanding—CZI is launching a 1,000 GPU computing cluster by June to facilitate training even more powerful, large-scale models.
A Primer on the Next Generation of Antibodies [Abhishaike Mahajan, Abhishaike’s Substack, April 2024]
Antibodies have long been a cornerstone of immunotherapy, praised for their role in the adaptive immune systems of multicellular organisms. However, they come with their complexities and limitations, especially when adapted for medical use, which demands scalability and a focus on both short-term and long-term patient outcomes. As our understanding of biology advances, scientists are beginning to explore beyond traditional antibodies to develop more efficient therapeutic alternatives.
This blog post delves into the current landscape of antibody alternatives and what might lie ahead in the next decade. Here’s a brief overview of three promising developments:
1. Single-chain variable fragments (scFv): Although they're an older entry in the field of antibody engineering, scFvs are still relatively nascent. With nine drugs already on the market, scFvs represent a simpler, more manageable alternative to full-length antibodies.
2. Nanobodies: These are perhaps the most exciting development in antibody alternatives currently. Despite only one drug commercially available so far, nanobodies offer significant potential due to their small size and unique capabilities.
3. Antibody mimetics: Looking to the future, antibody mimetics are poised to redefine therapeutic approaches. These synthetic molecules can be designed to mimic antibody functions but are often easier to produce and can be customized to better meet clinical needs.
The focus on scFvs, nanobodies, and antibody mimetics reflects a shift towards alternatives that sit in a clinical gray area—well-explored academically but with impacts that are not yet fully understood in medical practice. Unlike their more established or nascent counterparts, these alternatives promise a balance of innovation and practicality, potentially leading to breakthroughs in how we treat diseases.
As we continue to push the boundaries of biomedical research, it's clear that the future will likely hold a shift from traditional antibodies to these more adaptable and possibly more effective alternatives. The journey from bench to bedside is complex, but the evolution of antibody technologies is a thrilling prospect for the future of medicine.
The Biotech Startup Contraction Continues… And That’s A Good Thing [Bruce Booth, LifeSci, 2024]
In his latest post, Bruce Booth of Atlas Ventures discusses notable shifts in the biotech sector. After a surge in startup creation fueled by pandemic-era excitement, recent data from Pitchbook shows that the number of new biotech companies getting off the ground has significantly decreased. In the first quarter of 2024, only about 60 new biotech ventures secured their initial round of funding—the slowest pace in eight years, and a sharp drop from the peak in 2021.
However, this slowdown isn't necessarily bad news. It suggests a return to more prudent practices in the venture creation space. Here are a few reasons why this could be beneficial for the sector:
Quality over Quantity: Launching a successful biotech startup involves more than just great scientific ideas—it requires careful setup and the right decisions early on. A slower pace allows for more thorough vetting of new ventures, potentially leading to more robust companies.
Experienced Leadership: High-quality leadership is crucial, yet experienced leaders are rare. Fewer new companies mean these valuable resources can be concentrated where they are most needed, increasing the chances of success.
Less Competition: The biotech field has seen a glut of companies vying for attention in the same therapeutic areas, which can lead to inefficiencies. With fewer companies being launched, there could be less overlap and more focused innovation.
Better Prospects for New Ventures: The anticipated future market conditions for selling or going public with these startups look more favorable now, with potentially less competition for mergers and initial public offerings.
However, it's essential to remember that the ultimate goal is to bring new medicines to patients. With fewer startups, some innovative treatments might remain undeveloped, stuck in the lab without the opportunity to reach clinical trials. While the current trend might be healthier for the biotech industry's economic aspects, it also poses a risk of slowing medical advancement.
Beyond Steel Tanks [Asimov Press, March 2024]
This short essay explores the current status, challenges and opportunities in biomanufacturing. One of the largest applications of biomanufacturing is the production of complex biomolecules that serve as many therapeutics today, including mRNA vaccines and GLP-1 obesity drugs such as Zepbound. Especially with the rising demand for such obesity drugs, their manufacturers Novo Nordisk and Eli Lilly are struggling to rapidly scale up production capacity.Â
There has been little radical innovation in biomanufacturing processes since Genentech first applied the process to the production of insulin in the 1980s. There are two paths forward to improve on existing processes: (1) Improving the current paradigm (2) exploring alternatives
Improving the paradigm: currently, complex biomolecules are manufactured by genetically engineering bacteria/yeast to produce a specific molecule, growing them in steel tanks and then purifying the chemical. At each stage, innovations can be made to increase purity, efficiency and yield.
Exploring alternatives: two promising areas are cell-free extracts and in vivo approaches. Cell-free extracts are touted to be more efficient since they only focus on the machinery that produces the compounds. In vivo approaches are based on the fact that native living things are the most efficient producers of biomolecules. A key example is the in vivo production of mRNA vaccines in the cell.
Moderna turns to AI to change how its employees work [Ned Pagliarulo, Biopharma Dive, April 2024]
Moderna is expanding its partnership with OpenAI to give its thousands of employees access to custom-built AI chatbots. Moderna already states that they have deployed more than 750 transformer models, including ones that help select vaccine doses for testing and for its lawyers to scan contracts. Amgen and Genmab are both working with OpenAI in a similar fashion.
Academic papers
Development of supramolecular anticoagulants with on-demand reversibility [Dockerill et al., Nature Biotech, April 2024]
Why it matters: A new platform for creating potent, selective, and rapidly reversible therapeutic inhibitors using a supramolecular strategy. By linking cooperative ligands with a reversible tether, on-demand termination of drug action can be achieved via a specific antidote. This provides a mechanism to potentially improve the efficacy and safety of drugs for anticoagulation and other important therapeutic applications.Nicolas Winssinger’s lab at the University of Geneva has invented a new method for the rapid reversal of drug activity on demand. The group designed a novel supramolecular drug platform that enables potent and selective inhibition of thrombin, a key enzyme in the blood clotting cascade, while also allowing for rapid reversal of anticoagulant activity on demand. Anticoagulants are important for preventing and treating thrombosis, but carry risks of adverse bleeding events. Current reversal strategies for anticoagulants have limitations in terms of cost, specificity, and effectiveness.
The supramolecular inhibitors consist of two cooperative ligands - one targeting thrombin's active site and the other its exosite II - linked together by complementary peptide nucleic acid (PNA) strands. The individual ligands have low potency and selectivity on their own, but assemble into highly potent and selective thrombin inhibitors when templated by the protein. Importantly, inhibition can be rapidly reversed by addition of a PNA antidote strand that outcompetes the PNA linker. The supramolecular anticoagulant showed potent activity in human and mouse plasma clotting assays, and a mouse thrombosis model in vivo. Impressively, in vivo anticoagulation was effectively reversed by the PNA antidote.
This supramolecular inhibitor approach provides a general mechanism to enable both high potency and selectivity as well as on-demand reversal of therapeutic activity. It could help improve the efficacy and safety of anticoagulant drugs. More broadly, this strategy of linking cooperative pharmacophores with reversible supramolecular tethers like PNA may be applicable to many other therapeutic targets beyond thrombin. The ability to rapidly switch drug activity on and off in a programmed manner opens up exciting new possibilities for making safer and more responsive medicines.
Improving microbial phylogeny with citizen science within a mass-market video game [Nature Biotechnology, 2024]
Why it matters: The microbiome’s importance has been implicated in a multitude of health issues, disease and also touted for use in synthetic biology applications. However, mapping the genetic diversity is tedious and requires the comparison and matching of millions of DNA sequences to existing databases. Citizen science, specifically the gamification of scientific problems, has emerged over the past decade as a promising approach to help with such large computational tasks. Instead of gamifying a task and keeping it as close as possible to its scientific framework, Borderlands Science has introduced a citizen science task into an already popular commercial video game to encourage further adoption and player retention. Since its launch in April 2020, Borderlands Science has engaged more than 4 million players who contributed with 135 million solutions to build a multiple sequence alignment that improves microbial phylogeny estimations and associations.
The sequencing data are processed to generate puzzles that are played in the Borderlands Science mini-game, which mimics a tile-matching game. The puzzles are made of four different types of bricks representing the four different nucleotides. Each column is a DNA fragment, and players must align the colored bricks to the guide columns on the left by inserting matching yellow tiles. Then the solutions are collected and aggregated to build an alignment of 1 million RNA sequences, which is used to build phylogenies that are benchmarked against other methods and reference trees.
An esophagus cell atlas reveals dynamic rewiring during active eosinophilic esophagitis and remission [Ding et al., Nature Communications, April 2024]
Why it matters: This research is significant as it provides the most comprehensive map of cellular interactions and alterations in the esophagus during both active EoE and its remission. Understanding these interactions is crucial as EoE is a complex disease driven by type 2 cytokine inflammation, commonly linked to food allergies and environmental factors, with potential severe effects like esophageal scarring and food impaction. By mapping cellular changes, the study not only enhances our understanding of EoE's pathogenesis but also identifies potential new therapeutic targets. Such insights are vital for developing more effective treatments that could prevent the disease's progression rather than merely managing symptoms.The study conducted by Jiarui Ding and colleagues presents an extensive cell atlas of the esophagus, highlighting dynamic cellular changes during active eosinophilic esophagitis (EoE) and its remission. The researchers analyzed over 421,000 single-cell profiles from esophageal biopsies of 22 individuals, identifying 60 distinct cell subsets. Key findings include an increase in ALOX15+ macrophages and PRDM16+ dendritic cells in active EoE, both expressing EoE-associated risk genes. Furthermore, the study discovered unique cell interactions and signaling pathways that could guide future therapeutic interventions targeting specific cell types beyond merely depleting eosinophils.
Notable Deals
Mammoth Biosciences and Regeneron partner to develop in vivo gene editing therapies that can target tissues beyond the liver. Mammoth will receive $100M in upfront and equity from Regeneron to jointly select and research collaboration targets, after which Regeneron will take over development. Regeneron has been developing methods using antibodies to improve targeting of delivery viruses to tissues aside from the liver. Mammoth has developed much smaller Cas9 systems that could fit inside such delivery vehicles, providing complimentary skills from each party.
Bristol Myers buys into Repertoire's autoimmune vaccines - the Flagship start-up will receive $65M upfront to create tolerizing vaccines that aim to rebalance a patients overactive immune system in three undisclosed autoimmune diseases.Â
Novartis pays PeptiDream $180M upfront to expand their work on peptides for conjugation with radio-therapies. The Swiss pharma giant already has a strong track record in the exploding radiopharma space with its blockbuster therapies Pluvicto and Lutathera
BigHat Biosciences collaborates with J&J to find antibodies against a suite of neurological targets
Enlaza raises $100M for covalent protein therapeutics. The approach introduces a proprietary unnatural amino acid that promotes covalency to the disease target to ensure a long-lasting outcome
What we listened to
What we liked on socials channels
Events
Machine Learning for Drug Discovery, Monday, 6th May 2024 - Vienna, Austria - before ICLR - https://mlfordd.com/
Field Trip
Did we miss anything? Would you like to contribute to Decoding Bio by writing a guest post? Drop us a note here or chat with us on Twitter: @ameekapadia @ketanyerneni @morgancheatham @pablolubroth @patricksmalone