- PLA gives overlooked biological evidence its own analysis before combining the findings.
- Published HCC research connects PLA-guided investigation with a four-gene survival signature validated in an independent patient cohort.
- As Anthropic and OpenAI expand biological discovery with AI agents, Oncotelic is targeting a 100-billion-vector protein program.
Oncotelic Therapeutics Inc. (OTCQB: OTLC) is building its PDAOAI drug discovery engine around a compelling premise: a clue buried in a small corner of the literature can redirect an entire research program. Its Partitioned Literature Aggregation (“PLA”) algorithm is designed to uncover those clues. A liver-cancer study published in Biomedicines on October 23, 2026, now provides a concrete example of that approach reaching patient-data validation.
In its October 26 announcement, Oncotelic highlighted the HCC publication and its connection to the broader PDAOAI discovery program, following the company’s October 22 keynote at the 22nd Drug Discovery Strategic Summit.
Giving the overlooked clue a chance
PLA partitions vectorized biomedical literature by disease or mechanism, analyzes each group independently, and aggregates the findings. A single similarity-ranked evidence pool can let a large research field crowd out a smaller but informative one. PLA gives those smaller bodies of evidence dedicated attention, preserving differences and exceptions that can point to a new biological question.
The platform’s intellectual-property foundation includes Oncotelic’s published international patent application WO 2025/117898 A1, Structures and methods for autonomous agents. The filing describes vector-based retrieval, separate document or fragment queries, and aggregation of the resulting responses.
PDAOAI currently indexes approximately 29 million PubMed abstracts. The opportunity is to connect evidence across that literature and turn an overlooked observation into a focused investigation.
Published HCC findings connect the method to patient outcomes
The HCC study applied PLA across 24 cancer-specific literature collections, bringing the contrast between successful mTOR-inhibitor use in renal cancer and unsuccessful historical trials in liver cancer into focus. Transcriptomic survival analysis then identified a four-gene composite: PPARGC1A, EIF4E, RRAGC, and FKBP1A.
In the discovery analysis, patients with high adverse-risk scores had a 2.59-fold higher hazard of death. The same score subsequently separated survival outcomes in an independent Japanese cohort of 203 patients, without selecting new genes or refitting its coefficients. This independently validated prognostic association provides a basis for identifying higher-risk patients and prospectively testing whether the signature can help guide mTOR-directed treatment research.
Anthropic and OpenAI reinforce the case for focused investigation
The broader AI field is producing evidence for dividing complex biological questions into focused investigations and combining the results. Oncotelic sees this convergence as independent support for the partition-and-aggregate principle behind PLA.
Anthropic’s enzyme-discovery campaign searched a database of approximately 1.94 billion protein clusters. Its agents narrowed the search, investigated individual families, and surfaced an unusual enzyme system with CRISPR-like repeats. The campaign used 949 agent sessions over 21.5 hours, with human scientists performing the laboratory follow-up. The work illustrates how targeted investigation can uncover an unexpected feature inside an enormous biological search space.
OpenAI, PerturbAI, and Scripps Research reported a complementary result on October 21. Thousands of agents examined gene disruptions in mouse-brain data, connected the results with literature, and combined their findings. At the largest tested workload, distributed agents cited more than 700 references versus 11 for a single agent under fixed model and resource conditions. In blinded comparisons, experts preferred the distributed reports 75% of the time.
The next ambition is 100 billion protein vectors
Oncotelic has announced plans to extend PDAOAI into genetic and protein discovery at a target scale of approximately 100 billion vectors. The goal is to carry focused biological investigation beyond published abstracts and into protein representations, using coordinated AI agents to pursue questions at a much larger scale.
The 100-billion figure is a development target for numerical representations of biological information. Anthropic’s 1.94-billion figure counts protein clusters, so the two describe different kinds of scale. For Oncotelic, the promise is a richer search space in which disease evidence, protein relationships, and testable hypotheses can be connected.
The strategic opportunity resembles a scientific land grab: finding valuable biological mechanisms and establishing what they can do before they are widely understood. Turning those findings into useful tools or treatments is the prize. Oncotelic enters that race with a PLA-based discovery engine, published HCC findings, and an ambitious next step. The next valuable clue may already be in the data.
For more information, visit PDAOAI.com and Oncotelic.com.
NOTE TO INVESTORS: The latest news and updates relating to OTLC are available in the company’s newsroom at https://nnw.fm/OTLC
About NetworkNewsWire
NetworkNewsWire (“NNW”) is a specialized communications platform with a focus on financial news and content distribution for private and public companies and the investment community. It is one of 75+ brands within the Dynamic Brand Portfolio @ IBN that delivers: (1) access to a vast network of wire solutions via InvestorWire to efficiently and effectively reach a myriad of target markets, demographics and diverse industries; (2) article and editorial syndication to 5,000+ outlets; (3) enhanced press release enhancement to ensure maximum impact; (4) social media distribution via IBN to millions of social media followers; and (5) a full array of tailored corporate communications solutions. With broad reach and a seasoned team of contributing journalists and writers, NNW is uniquely positioned to best serve private and public companies that want to reach a wide audience of investors, influencers, consumers, journalists and the general public. By cutting through the overload of information in today’s market, NNW brings its clients unparalleled recognition and brand awareness. NNW is where breaking news, insightful content and actionable information converge.
To receive SMS text alerts from NetworkNewsWire, text “STOCKS” to 888-902-4192 (U.S. Mobile Phones Only)
For more information, please visit https://www.NetworkNewsWire.com
Please see full terms of use and disclaimers on the NetworkNewsWire website applicable to all content provided by NNW, wherever published or re-published: https://www.NetworkNewsWire.com/Disclaimer
NetworkNewsWire
Austin, Texas
www.NetworkNewsWire.com
512.354.7000 Office
[email protected]
NetworkNewsWire is powered by IBN
