Therapeutics · Antibody discovery
Abalone Bio
An East Bay biotech that screens antibodies for what they do to a receptor, not how tightly they stick to it. Its first two molecules are published and preclinical, and most of the money behind them came from federal grants.
Scissors by the blades
There is a picture on Abalone Bio’s website of a hand gripping a pair of scissors by the blades. Richard Yu, co-founder and chief executive, uses it to explain what he thinks the antibody field measures wrong.
“Yeah, that’s a super tight interaction, but, like, that’s not gonna help you cut paper,” he told the Biotech Startups Podcast in 2026.
Antibody discovery optimizes for binding affinity because affinity is cheap to score. Hold a receptor tightly enough and you can block it. Switching one on is a different problem: the antibody has to grab in one place, in one way, and move a protein that is jiggling inside a membrane.
Abalone’s answer is to score the outcome instead of the grip. Its FAST platform, for Functional Antibody Selection Technology, puts a human receptor into engineered yeast and wires the receptor’s signaling to whether the cell survives. Antibodies that activate the target keep their host cell alive, and sequencing reads out which ones did it. The company says it measures antibodies “100 million at a time, 100X+ the throughput of others.”
The hardest target class for antibodies
G protein coupled receptors sit at the center of human physiology and at the edge of what drug developers can reach. In its own NSF outcomes report, Abalone put it plainly: roughly 30 percent of approved drugs target GPCRs, only 27 percent of GPCRs have been drugged, and those drugs are almost all small molecules or peptides.
The first target, cannabinoid receptor 2, shows why antibodies are attractive and why they are rare. CB2 damps inflammation in macrophages, a long standing route to fibrosis. It also looks very much like CB1, the receptor in the brain that produces psychoactive effects, so small molecules keep hitting both. Antibodies are specific enough to tell them apart and, as a rule, stay out of the brain.
Two friends from the Molecular Sciences Institute
Yu took a biophysics and computer science degree at UC Berkeley in 1993 and a PhD at Yale in 2000, then joined the Molecular Sciences Institute, where Roger Brent and Sydney Brenner studied how cells process information. He met Carlos Gustavo Pesce there in 2003, an Argentine biologist who had worked with Peter Walter at UCSF before turning to synthetic biology.
They co-founded an algae biofuel company in 2008. Yu later ran the QB3 incubator and was a principal at Mission Bay Capital, sitting through hundreds of pitch meetings before making one of his own. Pesce went to Bolt Threads and Amyris. In 2018 they started again, on antibodies. The company is named for the shellfish Pesce, an open ocean swimmer, used to dive for in Northern California.
March 9, and then February
Their Y Combinator Demo Day was set for March 19, 2020. It was moved up to March 9 and taken online as the pandemic closed in, one slide each. The company then took a suite in an Emeryville business park that had been a Pixar sound studio, and Yu, who had built lab space for a living, turned the sound rooms into lab rooms so scientists could work apart.
Eleven months later, in February 2021, Pesce died after a ski accident. He was 53. Yu was at his computer with a grant deadline a week away when the message came from Pesce’s wife.
The team held. Toshihiko Takeuchi, a former XOMA discovery executive who had been advising the company, came on as chief scientific officer. Monica Schwartz and Lauren Schwimmer were already there, and both appear in the federal record as principal investigators: Schwartz on the NSF Phase II, Schwimmer on six of the nine NIH awards.
Grants first, partners later
Public money arrived before any partnership. NSF gave a $225,000 SBIR Phase I in January 2018, with Pesce as principal investigator, then a Phase II that ran to the end of 2023 and drew $1,599,997. NIH added nine awards worth $5,620,424 between FY2019 and FY2024, across neuropathy, liver fibrosis, bowel disease, endometriosis pain and a peripherally restricted opioid agonist.
The deals came later: Penn in March 2025, Sihuan Pharmaceutical in October 2025, Pfizer in January 2026. The Penn work is the most specific. Penn runs one of the few labs with a colony of Asian musk shrews, a small mammal that can vomit, which makes it a better early test than a rodent for the nausea that drives about half of GLP-1 patients off treatment within a year.
What is proven, and what is still claimed
| Evidence | What the record shows | Source type |
|---|---|---|
| Peer-reviewed data | Molecular Pharmacology, March 2026 (PMID 41740200), by Abalone Bio and Mount Sinai authors. AB120 and AB150 suppress inflammatory cytokines in CB2 cells, do nothing in CB1 cells, and cut collagen expression in precision cut human liver slices. In vitro and ex vivo only. | Public record |
| Preprint | bioRxiv, December 1, 2025 (PMID 41383775): CB2 antibody agonists for chemotherapy induced peripheral neuropathy. Not peer reviewed at that stage. | Public record |
| Federal grants | NSF $225,000 (2018) and $1,599,997 (2019 to 2023). NIH $5,620,424 across nine actions and five project lines, FY2019 to FY2024. Total $7,445,421, which supports the company figure of “$7M in non-dilutive grant funding.” | Public record |
| Patent | A granted US patent for displaying antibodies in the yeast periplasmic space, filed December 10, 2018, issued October 24, 2023. Inventors: Yu, Pesce, Rodrigo Baltanas, Brett Robison. | Public record |
| Clinical trials | No registered studies and no FDA records as of September 22, 2026. Expected of a preclinical company, and still the gap. | Not found |
| Partnerships | Penn (March 2025) carries “an exclusive option agreement covering key intellectual property developed through the partnership.” Sihuan (October 2025) has a “multi-million dollar upfront payment.” Pfizer (January 2026): “All financial and operational details are confidential.” | Company-stated |
| Partnership economics | The YC profile claims “3 partnerships with $3M in revenue and $125M in downstream value.” The Sihuan release says that deal alone carries payments that “have the potential to surpass $1 billion.” No partner filing confirms either. | Claims differ |
| Equity raised | One Form D touches the company, and it belongs to an outside vehicle: $165,375 from six investors, first sale November 14, 2022. Abalone Bio, Inc. has filed none of its own. | Public record |
| Superlatives | Releases call Abalone Bio “the only company measuring functional activity data at scale” and claim agonists for “2 out of the 8 G-protein coupled receptors ever drugged by biotech.” Neither has an independent check. Drug Discovery News does report that “only a handful of GPCR agonist antibodies have been reported in the scientific literature.” | Company-stated |
Read plainly: the science has cleared a real bar. A peer-reviewed journal and an independent Mount Sinai laboratory have put their names to antibodies that switch on a receptor small molecules could never hit cleanly, and the work reached human liver tissue. Everything after that is still ahead. No animal efficacy paper for the obesity program, no development candidate, no trial, and no disclosed dollar value on any of the three deals. The business claims run ahead of the documents.
What to watch
- A named development candidate and an IND timeline for the metabolic program.
- Published in vivo data from the Penn work, especially the musk shrew nausea model.
- Whether the Pfizer collaboration converts into a licence with disclosed economics, or quietly ends.
- A Form D or an announced round filed under Abalone Bio, Inc. itself. There has never been one.
- NIH renewals. Federal money here is 75 percent NIH, and the largest project line ran out in July 2025.
In their words
“Yeah, that’s a super tight interaction, but, like, that’s not gonna help you cut paper.”
Richard Yu, co-founder and CEO, Biotech Startups Podcast, 2026 ·Interview
“FAST identifies special antibodies that recognize rare, specific conformations of the receptor that traditional methods would have simply overlooked.”
Richard Yu, to Drug Discovery News, March 2026 · Independent
“Most importantly, we moved beyond the petri dish and demonstrated that these antibodies work in human liver tissue.”
Scott Friedman, Mount Sinai, to Drug Discovery News, 2026 · Independent
“For years, the missing piece in obesity therapeutics has been the ability to selectively target the right neurons without affecting others.”
Matthew R. Hayes, PhD, University of Pennsylvania, March 2025 release · Partner release
“We believe this collaboration could unlock new approaches to potential treatments for diseases with high unmet need, by targeting biology in a way that has long been considered intractable.”
Richard Yu, January 2026 Pfizer release · Company release
“Fibrosis is the common denominator in many chronic diseases, especially liver disease, which is responsible for roughly four percent of deaths globally.”
Richard Yu, to Drug Discovery News, March 2026 · Independent
“Abalone Bio carries on the legacy of co-founder Gustavo Pesce.”
Abalone Bio about page, accessed September 2026 · Company
Related companies
Sources
- Public recordCannabinoid receptor 2 activating antibodies, PMID 41740200
- Public recordCB2 antibody agonists for neuropathy, PMID 41383775
- Public recordNIH RePORTER awards to Abalone Bio, Inc., nine actions
- Public recordNSF awards 1747391 and 1853147, with outcomes reports
- Public recordForm D, Abalone Bio a series of Sharding Holdings Management LLC
- Public recordUS patent on yeast periplasmic antibody display
- IndependentAI-designed antibodies crack a long-elusive target for fibrosis
- IndependentPartnership to discover GPCR-modulating antibodies
- IndependentEast Bay company aims to tame a shrew
- InterviewRichard Yu, parts 1 to 4
- CompanyPfizer research collaboration release
- CompanyMount Sinai publication release
- CompanySihuan Pharmaceutical collaboration release
- CompanyUniversity of Pennsylvania collaboration release
- CompanyAbalone Bio website, six pages
- CompanyAbalone Bio profile and founder bio
Profile researched and written by Healthcare Discovery. Last updated September 29, 2026.
