“Evo is a genomic foundation model that enables prediction and generation tasks from the molecular to genome scale. Using an architecture based on advances in deep signal processing, Evo is trained on 7 billion parameters with a context length of 131 kilobases at single-nucleotide resolution. Evo captures two fundamental aspects of biology—the multimodality of the central dogma and the multiscale nature of evolution.“
China is making immense progress in many fields. Biotechnology is no exception.
“A 2024 IQVIA report shows that the share of clinical trials launched by Chinese-headquartered biopharmaceutical companies rose from 3 percent in 2013 to 28 percent in 2023, suggesting a growing involvement of Chinese companies in early-phase drug development.“
This looks like a big deal – “The bridge RNA system is a fundamentally new mechanism for genome design,” said Dr. Patrick Hsu, senior author of the study and an Arc Institute Core Investigator and University of California, Berkeley Assistant Professor of Bioengineering. “Bridge recombination can universally modify genetic material through sequence-specific insertion, excision, inversion, and more, enabling a word processor for the living genome beyond CRISPR.“
Full Arc Institute blog here and AI simplified version here.
“According to the latest Pitchbook data, venture creation in biotech hit its slowest quarterly pace in eight years during 1Q 2024. With just over 60 new biotechs raising their first round of financing, the sector’s company formation activity has slowed 50-60% from its historic peak in 2021.“
The piece argues why this is a positive for the health of the sector.
There is probably enough evidence now to suggest GLP-1s, like Ozempic, work to treat all kinds of addiction.
Why does this matter? “Addiction kills more Americans than cancer or heart disease but only 4% of people with substance use disorders currently receive medication.“
Tolerability looks to be important long-term followed closely by supply, the key constraint currently as GLP-1 molecules are peptides, which could be solved instantly if a non-peptide white pill is developed.
Surprisingly a flurry of deals at the end of the year means 2023 was a record in terms of the number of deals >$1bn in value (source: Centerview, h/t Tema).
One of the reasons is likely a looming patent cliff.
Tema estimates that large-cap pharma has $838bn of dry powder – enough, even after the rally, to buy almost 70% of the XBI (without a premium).
Healthcare is one area where the application of AI, in its LLM and other forms, could be enormous.
This nice article from AlphaSense Expert Insights explores the topic, mirroring the huge rise in expert calls in the sector mentioning the term.
It is not all areas that can be bent to the will of ML. As this piece argues, academic literature and the correspondent knowledge graph is both difficult and not that useful to program.
If you want to read some of these transcripts, you can grab a two-week free trial.
More and more anecdotal evidence suggests that GLP-1 agonists (like semaglutide, the active ingredient in popular weight loss drugs Wegovy and Ozempic (off label)) could dampen cravings for alcohol.
A survey of individuals taking these drugs by Morgan Stanley has confirmed this phenomenon.
Up to six clinical trials are testing semaglutide in alcohol use disorder – which could provide conclusive proof.
Investors are generally trying to determine the implications of the weight-loss revolution. As this chart shows the number of expert calls done by investors on AlphaSense on the topic has risen rapidly.
Alpha Sense also has a handy guide to the Obesity space here (need to sign up).
What if drug development could one day be like engineering?
Artificial intelligence has the potential to make this a reality, as is so well explored in this accessible Forbes profile.
One of the most exciting parts they highlight is de novo protein design i.e. building proteins from nothing (this is a nice post about it from Derek Lowe).
“An analysis of biopharma pipelines shows that the average number of assets per target investigated has more than doubled: 3 assets/target in 2000 to 7/target today. The number of assets in clinical development is outpacing the number of biological approaches, particularly in oncology.”
This large (22m records) cross sub-discipline study found that nearly one in ten people (13% for women and 7% for men) have an autoimmune disease – much higher than historic estimates.
this “research also confirmed that some autoimmune diseases tend to cluster together (for example, one person with a first autoimmune disease is more likely to develop a second autoimmune disease than someone without autoimmune disease), however at a much larger scale and for a much larger set of autoimmune diseases than previous studies.“
One of the most useful tools in a healthcare investor’s arsenal is medical conferences.
Why? Key opinion leaders (KOLs) meet to discuss the latest data coming out from company-sponsored trials and research. Their views give an unrivaled and direct insight not just into the probability of success but also the commercial potential of various medical interventions.
One way to get a quick overview is summary quotes captured by the Stream blog like this summary of ACC ’23.
NB If you want to have free two-week trial access to the full database of transcripts click here.
