One assignee leads the ranked field with 49 records, well ahead of the fifth-place assignee at 18 and the tenth at 1, across the 10 companies the ranking covers. That leader's filings concentrate around barcoded combinatorial library methods and recursive multiplex engineering, which also happen to be the most-cited records in the dataset. A smaller cluster of university and biotech assignees fills out the rest of the ranking, several of them tied together through co-assignee filings rather than filing independently.

No — the documented trend shows filings falling from a peak of 16 in 2021 to 1 by 2024, a -94% drop over that span. That said, 2024 is the most recent year that can be treated as complete; patent publication typically lags filing by around 18 months, so 2025 and 2026 counts are still filling in and should not yet be read as confirmation of a permanent decline. Anyone tracking this space should revisit the trend once another 12–18 months of publications land.

This dataset was built around multiplex editing, recombineering and genome-scale engineering methods paired with the practical problems that come with editing at scale: off-target mutation, per-cycle editing overhead, selection marker recycling, cycle automation, genotype verification and strain stability. In classification terms, 88.0% of the 125 records carry a C12N microorganism/genetic-engineering class, with medicinal preparations (A61K) and peptide/protein claims (C07K) frequently layered on top. That combination distinguishes production-oriented editing patents from general genome-editing filings that do not address tracking or verification across cycles.

US20170369870A1, assigned to Inscripta, covers methods and compositions for trackable genetic variant libraries, recursive engineering, multiplex engineering, and enrichment for editing using nucleic acid-guided nucleases. It is one of a closely related cluster of barcoded-combinatorial-library filings that together account for the highest citation counts in this dataset, which signals it has shaped how later filings in the space are drafted. Anyone building a multiplexed, trackable editing workflow with barcode-based variant tracking should read the full claim set closely rather than relying on the abstract, since related filings in the same family extend similar claims across multiple jurisdictions.

The classification data points to smaller, more specific claim areas that sit underneath the dominant C12N core: automated genotype verification across editing cycles, selection marker recycling methods that avoid counter-selection steps, and off-target mutation detection built into a multiplex workflow. These appear as secondary themes within the 125 records rather than as standalone dense clusters, which suggests fewer directly competing claims. A first filing in one of these areas would need to tie the specific operational problem (recycling, verification, or detection) to a genome-scale method claim rather than claiming the detection or recycling step in isolation.