Pumpkinseed Aims for Full-Length Protein Sequencing Following $20M Series A Raise
The Stanford spinout could have proof-of-concept for 300 amino acid proteins within 12 to 18 months, CEO Jen Dionne said.
San Jose, California – Pumpkinseed, a Stanford spinout developing silicon chip-based protein sequencing, aims to demonstrate proof-of-concept for full-length 300-amino-acid proteins within 18 months — a milestone that could challenge competing approaches from mass spectrometry and nanopore-based methods.
“We may need to consider fragmentation and then stitch them back together. but I would suspect, within the next 12 to 18 months, we should be able to have good preliminary results showing proof of concept for full length proteins,” Pumpkinseed CEO and Cofounder Jen Dionne told Ion Genomics at the 2026 SynBioBeta conference.
To help them towards that goal, the Stanford University spinout raised $20 million in Series A financing, announced earlier this week. NfX and Future Ventures co-led the round, with participation from Base4, ADVentures, and Stanford University, among other investors.
Pumpkinseed has also received a contract from the Defense Advanced Research Projects Agency (DARPA) for an undisclosed amount that will aid development of full-length protein sequencing. “They are very interested in biological threat detection, so looking at non-canonical amino acids and native proteins in various environmental samples,” Dionne said.
In addition to working on its de-novo sequencing and identification of proteins with high-throughput Raman (deSIPHR) technology, Pumpkinseed plans to expand its biopharma and biosecurity partnerships and to advance artificial intelligence models built on its data.
The company is entering a crowded field. Mass spectrometry remains the standard for protein identification but struggles with full-length sequencing of large proteins. Several companies, including Oxford Nanopore Technologies and Portal Biotech, are developing nanopore-based approaches, while others such as Quantum-Si are exploring different single-molecule methods.
Founded by researchers in Dionne’s Stanford lab, Pumpkinseed’s technology uses advanced photonics to identify biomolecules, including peptides as well as nucleic acids . It uses features embedded on standard silicon chips to detect these molecules, offering applications in research and diagnostics. The company is able to manufacture chips with 100 million sensors per square cm and currently can sequence peptides 30 amino acids long.
Dionne suggested that their sequencers could one day analyze 10 billion letters per day. So that is our 100 million sensors per square centimeter, each sequencing a 100 length residue peptide within a 24 hour period,” Dionne said.
The startup has already partnered with Genenetech to use its technology to identify immunopeptides relevant in oncology.
The Raman scattering data generated by the firm’s devices and used to identify peptide sequences “is extremely information rich, not only in terms of primary structure, but also secondary and tertiary structure in terms of binding,” Dionne noted. This data is driving several pilot programs with companies building AI models of biology.
For now, the company is focusing on getting longer peptide sequences of its tech. “By the end of the year, we can probably get to 100-length residues,” Dionne said.