Most-used affinity proteomics platforms ranked by how many preprints say they used each
Affinity proteomics measures hundreds to thousands of proteins at once in a small sample, most often blood plasma. Antibodies or aptamers pick out each protein. Updated daily (as of Nov 16)
Olink is the most-used affinity proteomics platform.
Olink had 54% of the uses in Jun–Nov 2026, about the same as the year before. Alamar NULISA is second with 25%.
| # | Platform | Last 6 months | Growth | Share | Total | By year |
|---|---|---|---|---|---|---|
| 1 | 92 preprints | +30% | 54%same as a year earlier | 262 56% | ||
| 2 | 43 preprints | New | 25%up from 12% | 79 17% | ||
| 3 | 34 preprints | −22% | 20%down from 34% | 127 27% |
Last 6 months is Jun–Nov 2026. Growth compares it with Jun–Nov 2025, as a share of all preprints. Total counts every preprint since April 2025 that says it used the platform.
Top institutions for each platform
Karolinska Institutet12
Icahn School of Medicine at Mount Sinai7
University of Oxford5
University of California, San Francisco5
Stanford University4
University of Pittsburgh7
Emory University6
Stanford University4
Washington University in St. Louis3
University of Oxford3
Stanford University8
University of California, San Francisco5
Washington University in St. Louis5
University of California, San Diego5
Imperial College London4
Counts every preprint since April 2025, by the corresponding author's institution.
Olink is growing fastest.
Up 30% in Jun–Nov 2026 vs the previous year as a share of all preprints. SomaLogic SomaScan was down 22%. Alamar NULISA is new, in preprints since Q2 2025. All three together grew 30%. Compare side by side: Olink vs SomaScan →
Olink
NULISA
SomaScan
The last bar is the quarter so far.
Latest preprints using affinity proteomics
How we count which preprints use each platform
An AI model reads each preprint in full — its text, its tables and its supplementary files — and lists the products the authors used. A preprint counts for a platform when that record says the lab used it in this work, not when it only analyzed data others made with it or only mentioned it. Each preprint counts once, by its first version, if it was first posted on bioRxiv or medRxiv in April 2025 or later. We read new preprints within days of posting, so the newest ones (through Nov 16, 2026) are already counted, but those days are not complete yet. Six-month counts and growth figures stop at Nov 12, 2026, the last day nearly every preprint has been read, and charts draw a quarter that is still being counted in a paler color. A preprint that used two platforms counts for both. Each title links to the preprint on bioRxiv or medRxiv; we show no preprint text and no author names. Updated daily, most recently on Nov 16, 2026.