The Diffusion Index
Multi-Omics
36,995+ PubMed papers (with more uncounted — 2025–2026 hit PubMed’s search cap). “Multi-omics” went from a precise methodology term to a universal descriptor in under a decade. The diffusion curve records that transformation.
Papers analyzed (min.)
37k+
Peak tier 1+2 (2015)
184/1,000
Current (2026)
59.7/1,000
2026 papers (half year)
~10k
How many papers, and where they land
Annual multi-omics papers split into three tiers — Nature / Science / Cell, the rest of the top-tier specialist journals (Tier 1+2), and everything else. The log view keeps all three visible; the widening gap is the dilution the diffusion curve below captures as a ratio.
Log scale — each line is a count, so all three tiers stay visible despite spanning four orders of magnitude. The gap between the lines is the dilution: top-3 output barely moves while total volume explodes. 2006–2025 (2026 partial year omitted).
Peak: 98 per 1,000 in 2015 · current (2026): 31.8 per 1,000 — and still falling
Not a technology — a methodology trend
Multi-omics was never a single instrument or platform. It describes the practice of combining two or more omic layers — typically genomics with transcriptomics, proteomics, or metabolomics — to get a more complete biological picture. This is why it diffused differently than NGS or single-cell: there was no “you bought the box” moment, just incremental adoption as data integration tools matured and sequencing costs fell. The tier 1+2 score peaked early (2015–2018) when the approach was genuinely novel, then declined as it became expected practice.
The COVID inflection point
From 2019 to 2021, multi-omics paper volume exploded from 600 to 2,183 per year — a 3.6× jump driven substantially by COVID-19 research. The SARS-CoV-2 pandemic triggered a wave of papers combining transcriptomics, proteomics, and metabolomics to understand host response, identify drug targets, and characterize long COVID. Platform detection shows metabolomics (6,195 mentions) and TCGA multi-omics cancer studies (2,290) as the two dominant application areas throughout the field’s history.
When a term loses signal
By 2024, “multi-omics” was appearing in 6,000+ papers per year. By mid-2026, that pace annualizes to roughly 22,000 — faster than the term’s entire first decade combined. The score has collapsed accordingly: from ~184 per 1,000 (tier 1+2) in 2015 to ~60 today. This is linguistic commoditization — when a methodology word becomes a genus marker rather than a differentiator, the novelty premium evaporates entirely. The word is now descriptive, not aspirational.
What’s driving volume in 2025–2026
Three forces explain the recent paper explosion: (1) large biobank studies — UK Biobank, All of Us, FinnGen — combining genomics, proteomics (Olink), and clinical data at population scale; (2) AI-native multi-omics models (foundation models trained on combined omics layers); (3) spatial multi-omics, which combines spatial transcriptomics with protein or epigenomic data in the same tissue slice. None of these carry the novelty premium the original multi-omics papers did. The field is now infrastructure.
Where the papers come from
Share of multi-omics papers by first-author affiliation, 2013–2026. Parsed from 36,368 affiliations.
Broad multi-omics shows the steepest national concentration in the entire index: from a US-led start, China crossed the US in 2020 and reached 63% by 2026. This tracks the term’s linguistic commoditization — as multi-omics became a default descriptor rather than a method, the volume increasingly came from China’s high-throughput publication engine.
Methodology: Papers fetched from PubMed matching “multi-omics” OR “multiomics” OR “multi-omic” OR “multiomic”. Diffusion score = top-tier papers ÷ total papers × 1,000. Default view shows Tier 1+2 journals (more informative than top-3 for this field). Journal tiers assigned locally using a curated list. Note: 2025 and 2026 totals are minimum estimates — PubMed’s search result ceiling (~10,000 per query) was reached for both years, meaning the true paper count is higher and the true diffusion score is lower than shown. Data as of June 2026.