Medical devices & diagnostics · Proteomics tools

Correlia Biosystems

A UC Berkeley spinout that stopped waiting for proteins to drift toward an antibody and started pushing them there with an electric field. The instrument runs on one microliter of sample. Its evidence lives in conference posters.

Fishing with a net

A standard ELISA is a waiting game. Coat a well with antibody, add the sample, and let Brownian motion carry the target protein across a millimeter of liquid until it hits a capture site. Wash, add a detection antibody, wait again. Hours go by.

Correlia's cartridge replaces the well with a nanoporous polymer gel and replaces the waiting with a power supply. An electric field drives the sample through the gel, past capture sites packed close enough that a protein cannot miss them: five minutes of transport through a micro channel instead of hour long incubation in a plate. A trade blog put it in boats: one crew drops fishing poles and sits, the other drags nets. Correlia is the second crew, except the net stands still and the fish are pulled through it.

The payoff is volume. The instrument, called PIXI, needs about one microliter where a plate assay needs fifty to a hundred. In a mouse study that is the difference between a tail bleed and a dead animal at every timepoint. In needle biopsy work it is the difference between an assay and none.

Why the microliter is the product

Protein measurement has not had its sequencing moment. Plasma proteomics is crowded with platforms that each see a different slice of the proteome, and a September 2025 review in Communications Chemistry that ran eight of them against one patient cohort opened by noting that “comprehensive comparisons of the performance of different plasma proteomics platforms are limited.”

Correlia went after the older market underneath that one: the single target ELISA a pharma company runs thousands of times in preclinical work, measuring how much drug antibody is still circulating in a mouse on day seven. Y Combinator still lists the company with its early pitch, “Next-gen blood testing to provide 1000x more data to pharma companies”, and the federal SBIR profile describes an eventual future of “rapid, accurate and quantitative measurements from a drop of blood at the point-of-care”. Nothing in the record shows it attempted that second step. There is no FDA submission, no registered trial, and PIXI is sold as a research instrument.

Three engineers out of one lab

All three founders came from Amy Herr's bioengineering lab at UC Berkeley. Akwasi Apori earned his PhD there after a BS and MS in aeronautics and astronautics at MIT and a stint as a systems engineer at Boeing Satellite Systems. Samuel Tia earned his PhD in the same lab after building medical devices at Medtronic and RTI Surgical. M. Kursad Araz arrived as a postdoc with a Cornell PhD in MEMS and microfluidics.

The invention was not theirs alone. Berkeley's technology transfer office records that the company was founded to commercialize a protein immobilization technique invented in Herr's lab in collaboration with Novartis Diagnostics, and names Apori, Araz and Herr as the inventors. The underlying paper, by Araz, Apori, Salisbury and Herr, ran in Lab on a Chip on October 7, 2013. How it came free is the interesting part: according to a 2024 interview with Tia, the technology was “initially sponsored by a diagnostics company” and became licensable when that sponsor's internal group dissolved.

The founders picked it up the year the CITRIS Foundry opened at Berkeley, joined the first cohort, and spent six months writing grant proposals. It worked. NSF paid $150,000 in 2014. NIH paid $450,000 for a point of care sickle cell test and $223,898 in 2016 for a hepatitis C core antigen assay. The company moved through QB3, JLABS and MBC BioLabs, went through Y Combinator, and in March 2020 told CITRIS it had eight employees and “has raised $2.6 million in funding”.

From SimpleScan to PIXI

The first product had a different name. Federal and YC listings from the mid 2010s describe SimpleScan, an instrument for multiplexed panels of cytokines and antibodies. By 2023 the product was PIXI, single target rather than multiplexed at launch, built to sit on the deck of an Opentrons, Tecan or Hamilton liquid handler rather than replace it.

In between came the two events that define the public record. In January 2020 SLAS named Correlia its Ignite Award winner, picked from sixteen companies in its Innovation AveNEW startup area. In November 2020 the company announced a $7 million Series A led by Neotribe Ventures, with Cota Capital and the Regents of the University of California joining. The platform shipped in 2023.

Then the trail thins. The newest dated item on the company's own news page is June 1, 2024. No Form D has been filed since May 2023, and no independent article since May 2024. The site is still up and unchanged: it answered on September 23, 2026, and the Internet Archive's July 2026 capture matches the live page word for word.

What is proven, and what is still claimed

Read plainly: the physics is real, a university lab published it a decade ago, the federal government paid to develop it, and a product shipped. What has not happened is the step that turns a clever instrument into a standard. There is no peer-reviewed validation, no head to head study against a plate ELISA run by a party with nothing to gain, and no customer willing to attach its name. That is a common place for a small tools company three years after launch, and the point at which independent numbers begin to matter more than any claim on a website.

What to watch

• A peer-reviewed paper, or a poster whose authors all work somewhere else.
• The multiplex version. Launch coverage said multiplex assays would follow shortly; the product page still lists single target detection.
• A named pharma customer, or published pricing and placement figures.
• A third Form D. The May 2023 offering was left about $2.0 million short and never amended.

In their words

Correlia Biosystems funding profile on HVCFHealthcare Venture Capital Fund

Related companies

• BillionToOne: Shared investor: Neotribe Ventures. A Menlo Park laboratory that counts individual DNA molecules in blood, first to tell a pregnant patient whether her fetus is at risk for a recessive disease, now to read
• Moichor: Also in Diagnostics. An AI-assisted animal blood and cytology lab that started with birds and reptiles and now sells an in-clinic analyzer for dogs and cats
• InnaMed: Also in Diagnostics. A Philadelphia startup that built a microneedle, cartridge-based blood testing platform called HomeLab, sold multi-year testing services to pharmaceutical companies, and
• Instapath: Also in Diagnostics. A fluorescence microscope that turns a fresh, uncut biopsy into an image resembling a stained slide, in minutes

Sources

1. Public recordNIH RePORTER, R43HL123846 and R43DK109863NIH · FY2014 to FY2016
2. Public recordNSF SBIR Phase I award 1416633NSF · 2014
3. Public recordForm D filings, Correlia Biosystems, Inc., CIK 0001827990SEC · Oct 16, 2020 and May 9, 2023
4. Public recordMicrofluidic barcode assay for antibody-based confirmatory diagnosticsLab on a Chip · Oct 7, 2013
5. Public recordCompany record and inventor listUC Berkeley IPIRA · Sep 2026
6. IndependentSLAS Life Science and Technology Awards AnnouncedSLAS · 2020
7. IndependentCorrelia Launches PIXI ELISA PlatformClinical Lab Products · Apr 19, 2023
8. IndependentFishing with a Net, Not a PoleAnalytical Chemistry Startups · May 28, 2024
9. Interview7 Questions for Correlia Biosystems Co-Founder Akwasi AporiCITRIS · Mar 10, 2020
10. ConferenceEvaluating the PIXI Platform, poster T1330-02-10AAPS NBC · May 14, 2024
11. InvestorBBV Founder Spotlight: Correlia BiosystemsBlue Bear Ventures · 2022
12. CompanyCorrelia Biosystems Closes $7M Series APR Newswire · Nov 20, 2020