Computational Drug Design Patents: Who Leads, Where the Gaps Are 2026
- 59.1% concentration. The top five assignees among 2,574 records in scope hold 1,520 records combined — filing here is already dominated by a small group.
- Peptide and protein claims dominate. C07K appears on 62.4% of records and A61K on 59.6%, showing most filings anchor in composition-of-matter rather than pure computation.
- Bioinformatics stays a minority class. G16B and G06F sit at 5.9% and 5.0% of records respectively — the computational layer itself is still lightly claimed relative to the biological output.
Filing growth compares 2021 (132 records) with 2024 (91) — 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 2,574 records in scope (CR5), not by the ranked leaders only.
What this landscape covers
Computational drug design sits at the intersection of in silico molecular modelling and the downstream biological claims that make a candidate patentable — dosage composition, binding assays, therapeutic response, and delivery carriers. The search string used for this landscape deliberately pairs computational-method language with these outcome terms, so the corpus captures filings where the computation is tied to a therapeutic molecule or target-binding claim rather than software claimed in isolation.
That framing explains the IPC mix: peptide and protein chemistry (C07K) and medicinal preparations (A61K) lead by a wide margin, while bioinformatics (G16B) and digital data processing (G06F) trail well behind. The dataset spans 2015 through the 2026 cut-off, with 2017 as the peak filing year and the most recent years understated because publication lags filing by roughly eighteen months.
Let an AI agent run this analysis on your own technology
Pick a task. Every answer cites the patents behind it.
Filing trend and technology composition
Two views of the same 2,574-record corpus: how filing activity has moved year over year, and which IPC subclasses carry the claims.
Filing trend, 2017 peak to present
Filings peaked at 235 in 2017 and have declined since; 2021 to 2024 — the last year treatable as complete — fell 31%, from 132 to 91 records. Years after 2024 will keep filling in as publications catch up, so the apparent drop into 2025 and 2026 should not be read as the field slowing further.
IPC subclass composition
C07K (peptides & proteins) and A61K (medicinal preparations) each cover more than half of all records, with A61P (therapeutic activity) and C12N (microorganisms & genetic engineering) forming a secondary tier. G01N, C12Q, G16B and G06F — the analytical and computational classes — each sit under 15% of records, confirming that most granted claim value still attaches to the biological composition rather than the modelling method.
Shares are the percentage of the 2,574 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Computational Drug Design Patent Landscape with Eureka
This page is one run against one query. Ask Eureka your own question about computational drug design patent landscape and every answer comes back with the patent numbers behind it.
Try EurekaRepresentative and most-cited records
US20230335228A1 — Active Learning Using Coverage Score
A method for computational drug design includes defining a population of compounds, each with one or more molecular properties. From this population a training set with known biological properties is defined, and a subset of compounds outside the training set is selected. The method scores that subset based on the molecular properties of its members and evaluates it accordingly, with the score computed from the frequency of the molecular properties represented.Filed by Recursion Pharmaceuticals, published 2023-10-19 — an active-learning approach to selecting compound subsets for testing rather than a specific molecule claim.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | WO2015070083A1 | CRISPR-RELATED METHODS AND COMPOSITIONS WITH GOVERNING gRNAS | 507 |
| 2 | WO2013093809A1 | Engineered antibody constant regions for site-specific conjugation and methods and uses therefor | 308 |
| 3 | US20150232881A1 | CRISPR-RELATED METHODS AND COMPOSITIONS WITH GOVERNING gRNAS | 295 |
| 4 | US20160102322A1 | Crispr oligonucleotides and gene editing | 269 |
| 5 | WO2014082179A1 | Engineered immunoglobulin heavy chain-light chain pairs and uses thereof | 259 |
| 6 | US20070172483A1 | Methods for treating conditions associated with MASP-2 dependent complement activation | 235 |
| 7 | WO2009068204A1 | Anti-mesothelin antibodies and uses therefor | 195 |
| 8 | US20200362398A1 | Multiplexed signal amplification | 182 |
| 9 | US20090093024A1 | Methods of generating libraries and uses thereof | 156 |
| 10 | WO2015181805A1 | Modified antigen binding polypeptide constructs and uses thereof | 155 |
Citation counts favour older records simply by virtue of longer exposure — read them as a signal of influence within this searched corpus, not as a ranking of current technical importance.
Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
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 →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 →What the data means for filing strategy
Concentration, class distribution and citation patterns together point to where new claims still have room and where they don't.
Filing is already concentrated at the top
The top five assignees account for 1,520 of the 2,574 records in scope, and the top ten extend that to 70.5%. A long tail of single- and low-filing entrants sits behind a small group that has already staked out dense claim territory, particularly around antibody engineering and gene-editing constructs.
Composition claims outweigh computational-method claims
Peptide and protein chemistry and medicinal-preparation classes dominate the corpus, while bioinformatics and digital-processing classes remain under 6% each. That gap suggests the computational method itself is rarely the sole basis of a claim — it typically supports a biological composition or delivery claim instead.
Recent filing volume has eased from its 2017 peak
Filings dropped from 132 in 2021 to 91 in 2024, a real decline over a period with complete publication data. Combined with the 2017 peak of 235, the trend shows the field cooling from an earlier surge rather than accelerating, though 2025-2026 figures are still filling in and should not be read as confirmation of further slowdown.
Citation leaders skew toward gene-editing constructs
The most-cited records in this corpus are dominated by CRISPR-related compositions and engineered antibody constant regions, both filed earlier in the window. Their citation counts reflect years of exposure inside the corpus as much as ongoing technical centrality, a pattern typical of citation analysis in any searched patent set.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to computational drug design patent landscape, with the prior art for and against each one.
Assignee landscape and where the gaps sit
A ranked field of 100 companies shows a leader well ahead of the pack, a fast drop to a mid tier, and several technically specific branches with comparatively few claims filed against them.
One assignee sits well ahead of the ranked field
The leading assignee's 560 records dwarf the fifth-place figure of 111 and the tenth-place figure of 45 — a steep drop-off rather than a gradual one. Anyone entering this space should expect the leader's portfolio to anchor much of the prior art in core binding-assay and therapeutic-molecule claims.
Collaboration is limited but concentrated
Only ten co-assignee pairs appear in the dataset, and the strongest link — 507 shared records — sits far above the next pairs at 40 and 29. Most other collaboration in this field is comparatively thin, meaning joint-filing partnerships are the exception rather than the norm.
Recent-year activity has thinned across named assignees
Several previously active assignees show zero filings in the latest tracked year, with year-over-year changes of -100% for some, while only one assignee shows single-digit activity continuing. This is consistent with the broader -31% filing decline from 2021 to 2024 rather than isolated to one company.
| Assignee | Recent year | YoY |
|---|---|---|
| Bayer AG | 3 | — |
| The Broad Institute, Inc. | 1 | 0% |
| Omeros Corp | 0 | -100% |
| University of Leicester | 0 | -100% |
| Bayer Intellectual Property GmbH | 0 | — |
| Bayer Pharma AG | 0 | — |
| Caribou Biosciences, Inc. | 0 | -100% |
| MorphoSys AG | 0 | — |
Where to take this analysis
The dataset points to a concentrated core and a thinner computational layer around it. The next step is usually narrower: checking a specific claim against the leader's family, or scoping a filing in one of the lighter-claimed branches.
Run a freedom-to-operate check
Before drafting new claims in binding-assay or therapeutic-molecule space, check them against the leading assignee's family and the top-cited CRISPR and antibody-construct records.
Explore in Patsnap EurekaScope the under-claimed branches
Bioinformatics-linked target-binding methods and coverage-score-style active learning remain thin relative to the corpus; a first claim there faces less crowded prior art.
Explore in Patsnap EurekaCommon questions on computational drug design patents
It is highly concentrated at the top. Among the 2,574 records in scope, the top five assignees combined hold 1,520 records, or 59.1% of the field, and the top ten extend that to 70.5%. Beyond that tenth position, filing counts drop quickly into a long tail of companies and institutions with far fewer records each, so a new entrant should expect the leading handful to define the prior-art landscape for core claims.
In this corpus, bioinformatics (G16B) and digital data processing (G06F) classes each appear on under 6% of the 2,574 records, while peptide/protein chemistry (C07K) and medicinal preparations (A61K) appear on well over half. That distribution suggests claims built purely around the computational method, without an attached therapeutic molecule or composition, are comparatively rare and likely to face examiner scrutiny on subject-matter grounds as much as on prior art.
Publication of a patent application typically lags its filing date by around eighteen months, so any year within that window will always look under-filled compared to earlier years, purely as an artefact of the publication pipeline. The reliable comparison here is 2021 to 2024, which shows a real 31% decline from 132 to 91 filings. Figures for 2025 and 2026 will continue to rise as more applications publish, so they should not be read as confirmation of a further slowdown.
Analytical and computational IPC classes — G01N (material analysis), C12Q (enzyme/DNA measuring), G16B (bioinformatics) and G06F (digital processing) — each cover under 15% of the 2,574 records, well below the peptide-chemistry and medicinal-preparation classes that dominate the corpus. Branches like coverage-score-based active learning and bioinformatics-linked target-binding methods sit in this lighter-claimed space, making them a reasonable starting point for scoping a new filing with less prior art to navigate.
US20230335228A1, filed by Recursion Pharmaceuticals and published 2023-10-19, claims a method for selecting a subset of compounds from a larger population using a coverage score derived from the frequency of molecular properties in that subset, evaluated against a training set with known biological properties. It is a method for compound selection and evaluation rather than a claim over any specific molecule or therapeutic composition. Work that scores or selects compounds using a materially different metric, or that operates on a differently defined training set, has room to design around its specific claim language, though a full freedom-to-operate review of its claim set is advisable before relying on that distinction.
Research Computational Drug Design Patent Landscape in depth with Eureka
Go past this page: query the whole computational drug design patent landscape 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.