Fastest
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →https://www.patsnap.com/resources/blog/rd-blog/generative-protein-design-patent-landscape/ · Patsnap · data cut-off 2026-07-31 · downloaded from the live page
Patent Landscape · AI Protein DesignGenerative protein design patents: mapping the leaders, the filing curve and the open ground
- Concentrated at the top. the leading assignee alone accounts for 32 of 106 records, and the five leading assignees together hold 70.8% of all 106 records in scope.
- Filing kept climbing through the last complete year. activity rose from 10 records in 2021 to 14 in 2024, a 40% increase over that three-year span, before the most recent two years understate true filing due to publication lag.
- Peptide and bioinformatics classes dominate the claim space. C07K appears on 51.9% of records and G16B on 44.3%, while computational-chemistry class G16C sits at just 7.5% — a narrower, less-crowded route into the same problem.
106
Published Records
71%
Top-5 Share of All Records
+40%
Filing Growth 2021→2024
WO
Leading Jurisdiction
Filing growth compares 2021 (10 records) with 2024 (14) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 106 records in scope (CR5), not by the ranked leaders only.
Published byPatsnap Research··7 min readSourced from Patsnap Eureka
Overview
What the generative protein design patent record actually shows
The search set covers 106 published records spanning 2015 through the mid-2026 data cut-off, built around claim language for protein design models, de novo and diffusion-based design methods, paired with metrics like designability, sequence recovery, binder design and expression solubility. This is a young, university-anchored field: academic and research-institute assignees dominate the ranking, and the filing curve only started building meaningful volume from the early 2020s.
Filing activity is not evenly spread. A small number of assignees hold most of the recorded families, receiving-office activity concentrates heavily in PCT and US filings, and the underlying IPC classes show peptide chemistry and bioinformatics claims doing most of the work while computational-chemistry framing remains comparatively open.
Filing activity, 2017-2026
Source: Patsnap Eureka. Assignee ranking and totals. Derived from a Patsnap search on Generative Protein Design covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
10,000 free credits to start
The Numbers
Filing trend and technology composition
Two views of the same 106-record set: how filing volume has moved year over year, and which IPC subclasses the claims actually sit in.
A field still building volume
Filings moved from zero in 2017 to a peak of 21 in 2025, with the 2021-to-2024 span showing a documented 40% rise (10 to 14 records). Treat 2025 and 2026 as undercounts rather than a plateau or decline, since publication typically lags filing by around 18 months.
Where the claims sit
C07K (peptides and proteins) and G16B (bioinformatics) each appear on close to half of all records, reflecting that most filings pair a molecular composition claim with a computational method claim. A61K and C12N follow at meaningfully lower shares, and G16C (computational chemistry) and G06F (digital data processing) sit under 8%, marking a narrower claim style rather than an empty one.
Shares are the percentage of the 106 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Source: Patsnap Eureka. Filing trend and technology composition. Derived from a Patsnap search on Generative Protein Design covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.
Go deeper on Generative Protein Design with Eureka
This page is one run against one query. Ask Eureka your own question about generative protein design and every answer comes back with the patent numbers behind it.
Try EurekaRepresentative Filing
A representative claim from the leading edge
Representative record
Zinc finger design using a hierarchical machine learning model
THE GOVERNING COUNCIL OF THE UNIVERSITY OF TORONTO
A machine learning model for designing Zinc Finger proteins that bind to a given nucleic acid target sequence, using a hierarchical architecture: a first layer trained on single-helix specificity data across diverse interface environments, and a second layer trained on dual-helix specificity data.Filed by The Governing Council of the University of Toronto, published 2025-06-12.
Most-cited records in the search set
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20090081675A1 | Methods, compounds and systems for detecting a microorganism in a sample | 57 |
| 2 | WO2016005969A1 | Method of computational protein design | 11 |
| 3 | US20170206308A1 | Method of computational protein design | 9 |
| 4 | US20160304922A1 | Host Cell Modified to Produce 2-Pyrrolidone | 7 |
| 5 | CN116894472A | 用于预测多肽结合力的神经网络模型的训练方法和装置 | 6 |
| 6 | WO2023288191A1 | Multiple de novo designed protein binding proteins | 6 |
| 7 | WO2021242780A2 | Modular and generalizable biosensor platform based on de novo designed protein switches | 6 |
| 8 | US20230416726A1 | Scaffolding protein functional sites using deep learning | 5 |
| 9 | WO2011143414A1 | RECOMBINANTLY PRODUCED ANTIBODIES TARGETING ErbB SIGNALING MOLECULES AND METHODS OF USE THEREOF FOR THE DIAGN… | 5 |
| 10 | WO2025122885A1 | Anti-her2 associated antibody compositions designed by artificial intelligence and methods of use | 3 |
Citation counts reflect influence within this searched corpus and skew toward older filings; they are not a measure of current commercial importance.
Patent titles are shown in the language they were filed in, not translated, so that each record stays verifiable against the original filing — a translated title will not match in Eureka or in any national register. Each row carries its publication number; clicking a row searches Eureka by that number.
Source: Patsnap Eureka. Citation counts and representative records. Derived from a Patsnap search on Generative Protein Design covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Run it yourself
Put your own technology through the same analysis
Where to run it
For builders
MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →Insights
What the numbers mean for a filing decision
Three read-outs from the concentration, trend and classification data that matter for deciding where to file and what to watch.
Concentration
70.8% of 106 records
held by the top 5 assignees
A short list of institutions controls most of the recorded claim space
With the leading assignee alone at 32 records and the top five together at 75 of 106, most of the documented claim space in generative protein design sits with a handful of university and research-institute filers rather than a broad field of commercial entrants.
Top 10 combined reach 89.6% of records.
Momentum
+40%
2021 to 2024 filing growth
Volume built steadily through the last complete filing year
Filings rose from 10 records in 2021 to 14 in 2024. The 2025 peak of 21 records and the 2026 partial year should be read cautiously given the roughly 18-month lag between filing and publication, not as evidence of accelerating or slowing pace.
2025 is the peak year recorded so far, likely still incomplete.
Classification
51.9% vs 7.5%
C07K share vs G16C share
Peptide and bioinformatics claims dominate; computational chemistry stays thin
C07K and G16B together anchor the majority of filings, pairing molecular composition with computational method claims. G16C, the computational-chemistry class, appears on only 7.5% of records — a sign that claim strategies built purely around simulation or scoring methods remain comparatively rare.
Classes overlap; shares sum to more than 100% of the 106 records.
Eureka AI Agent
Looking for what nobody has claimed yet?
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to generative protein design, with the prior art for and against each one.
Source: Patsnap Eureka. Co-assignee relationships and derived observations. Derived from a Patsnap search on Generative Protein Design covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Players
Who is filing, and where the activity has gone quiet
The ranking covers 53 companies and institutions counted across the 106 records in scope. Recent-year momentum figures show several of the historically active filers recording zero output in the latest year, which is consistent with publication lag rather than a genuine stop in research.
Leader
32 records
of 106 in scope
One assignee holds close to a third of all recorded filings
The top-ranked assignee's 32 records set it well apart from the rest of the field; fifth place holds 7 and tenth place holds just 2, showing a steep drop-off rather than a gradual one.
Ranking is family-based, not raw document count.
Collaboration
10 pairs
documented co-assignee links
Cross-institution filing is real but limited to a few strong pairs
Ten co-assignee pairs appear in the data, with the strongest linking two university filers on 7 shared records — evidence of sustained joint research programmes rather than one-off co-filings.
Strongest pairs recur across multiple records.
Momentum
-100% YoY
for several leading filers
Several of the most active historic filers show zero output in the latest year
Multiple institutions with strong multi-year filing records, including the field's most co-active pairs, show 0 records in the latest tracked year and a -100% year-on-year change. Given the roughly 18-month publication lag, this reads as a reporting gap rather than an exit from the field.
Treat single-year drops to zero as data-lag artefacts, not withdrawals.
Recent-year filing momentum by assignee
| Assignee | Recent year | YoY |
|---|---|---|
| University of Washington | 0 | -100% |
| Yeda Research and Development Co. Ltd. | 0 | — |
| The Governing Council of the University of Toronto | 0 | -100% |
| New York University | 0 | -100% |
| European Molecular Biology Laboratory (EMBL) | 0 | -100% |
| The Regents of the University of California | 0 | — |
| Fox Chase Cancer Center | 0 | — |
| APTAMINO LLC | 0 | -100% |
Source: Patsnap Eureka. Assignee-level momentum. Derived from a Patsnap search on Generative Protein Design covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
What's Next
Where to take this next
The dataset points to a field still consolidating around a small set of institutional filers, with room in specific technical corners.
Check freedom-to-operate against the leading assignee's claim family
With one assignee holding close to a third of all records, a design-around review should start there before looking at the long tail.
Explore assignee claim scopeWatch the G16C and G06F corners for new entrants
Computational-chemistry and pure data-processing framings remain thinly claimed relative to C07K and G16B, and that is where new filing strategies are most likely to find open ground.
Track emerging filingsSource: Patsnap Eureka. Forward-looking reading of the same dataset. Derived from a Patsnap search on Generative Protein Design covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
FAQ
Common questions on the generative protein design patent landscape
The dataset's leading assignee holds 32 of the 106 records in scope, well ahead of the rest of the field. The top five assignees combined account for 70.8% of all 106 records, and the top ten reach 89.6%, meaning most of the documented claim space sits with a small group of university and research-institute filers rather than being spread across many commercial entrants. Anyone doing freedom-to-operate work in this space should look closely at that leading assignee's claim family first.
Filing volume grew from 10 records in 2021 to 14 in 2024, a documented 40% increase over that span, and this is the most reliable read because 2024 is the last year that can be treated as largely complete. The years after that, including the apparent 2025 peak of 21 records, should not be read as either accelerating growth or a slowdown, because publication typically lags actual filing by around 18 months. The honest statement is that the field grew through 2024 and current-year figures are still filling in.
Claims cluster heavily around two IPC subclasses: C07K, covering peptides and proteins, appears on 51.9% of the 106 records, and G16B, covering bioinformatics methods, appears on 44.3%. A61K (medicinal preparations) and C12N (microorganisms and genetic engineering) follow at lower but still meaningful shares. Narrower classes like G16C (computational chemistry) and G06F (digital data processing) sit under 8%, indicating that claims framed purely around computational methods are comparatively rare compared to those pairing a molecule with a design process.
The classification data points to computational-chemistry-anchored claims (G16C) and pure data-processing framings (G06F) as thinner relative to the dominant C07K and G16B combination, together covering under 8% of records each. Specific sub-areas worth checking include binder design for non-antibody scaffolds, expression solubility prediction as a standalone claim, and novelty assessment tooling for generated sequences. These are technical directions implied by the search terms and class distribution rather than areas confirmed empty, so a proper freedom-to-operate search is still required before filing.
The data shows 10 documented co-assignee pairs, with the strongest pairing appearing on 7 shared records between two university filers, and two other pairs each appearing on 3 shared records. This indicates that collaboration in this field tends to be concentrated in a few sustained institutional partnerships rather than being widespread across the ranked list. Several of these same collaborating institutions show zero output in the most recent tracked year, which is consistent with publication lag rather than the end of the collaboration.
Answers are grounded in the same dataset. Derived from a Patsnap search on Generative Protein Design covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish.Run this in Eureka MCP
Research Generative Protein Design in depth with Eureka
Go past this page: query the whole generative protein design corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company's registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.