Three Major Research Project Milestones:
LNP / OA Osteoarthritis / Spatial Transcriptomics Advancing Simultaneously
ResearchLinkAI's in-house projects make their collective debut—three distinct research pipelines have entered the "Submission Ready" or "Pending GPU Execution" stages respectively. Two of these papers target journals including Frontiers in Pharmacology (IF≈4) and Cell Death & Disease (IF≈9), demonstrating that AI-native research possesses stable output capabilities across three domains: bioinformatics, pharmacology, and spatial omics.
In May 2026, three major internal research projects at ResearchLinkAI simultaneously reached critical milestones. This marks the company's first achievement of parallel delivery across multiple pipelines, disciplines, and methodologies since its founding. The three projects are led by AI researchers specializing in bioinformatics, pharmacology, and CS-AI respectively. The entire execution process relies on zero external PhD-level manpower—all manuscript writing, data analysis, and figure creation are completed within an AI-native workflow, with external experts participating only in final peer review and quality assurance.
Below is the current status, key outputs, and next steps for each of the three projects.
Study on LNP Immune Activation Mechanism
Driven by Network Pharmacology + Molecular Docking
Key Outputs
Research Progress
The entire workflow—from target screening, network construction, and enrichment analysis to molecular docking—was completed using a localized AutoDock Vina + dockstring solution. After two iterations, the manuscript evolved from v1 to v2, resolving issues regarding mechanism diagram hierarchy and reference coverage.
External expert review comments have been incorporated, and revision notes are fully archived in manuscript/revision-notes.md.
Multi-Tissue PANoptosis Comparison in OA
Inflammatory Cell Death Pathway Study Based on Public GEO Data
Key Outputs
Research Progress
The project expanded from single-tissue to a multi-tissue comparative design (cartilage, synovium, meniscus, etc.), identifying key PANoptosis nodes via differential expression and enrichment analysis. The manuscript has completed its sixth iteration (v6), passing both ARS multi-perspective simulated review and external expert evaluation.
With a target journal IF≈9, this marks the first time a ResearchLinkAI in-house paper targets a publication of this caliber.
Cross-Tissue Transfer in Spatial Transcriptomics
Spatial Omics + Deep Transfer Learning
Key Outputs
Research Progress
The proposal phase has completed two iterations, adding the Mamba architecture as a comparative baseline. The AutoDL execution plan has been documented in autodl-execution-plan.md,
and we are currently awaiting GPU resource scheduling.
This project validates ResearchLinkAI's capabilities in "AI researching AI"—treating deep learning methods themselves as the subject of study.
Business Impact
The collective unveiling of these three major projects signifies three things for ResearchLinkAI's business landscape:
First, the output capability of AI-native research is validated.
From project initiation, literature review, data acquisition, statistical analysis, and figure creation to manuscript writing, the entire workflow is led by AI researchers, with external experts participating only in final quality review. This means the company can produce workloads that originally required teams of ten or more people with a single-digit headcount.
Second, the "multiple yields from one source" business model is exemplified.
The two submission-ready papers will simultaneously spawn invention patent and software copyright applications; the method implementation from the spatial transcriptomics project can be directly converted into open course materials and SaaS components—a single research investment supports four output lines: papers, patents, copyrights, and content.
Third, client-side service capabilities are endorsed.
The three major service categories in our Taobao store—"Bioinformatics Analysis," "Molecular Docking and Pharmacological Analysis," and "Algorithm Development"—are backed precisely by the toolchains and method libraries accumulated from these three projects. Clients paying for these services are purchasing capabilities that have been repeatedly validated by our in-house research.
Next Steps
-
LNP Project: Finalize Cover Letter and high-resolution figures; submit to Frontiers in Pharmacology in early June.
-
OA Osteoarthritis Project: Complete final figure unification and supplementary material organization; submit to Cell Death & Disease in mid-June.
-
Spatial Transcriptomics Project: Procure AutoDL GPU resources and initiate baseline training; expect first batch of experimental data by late June and formal manuscript drafting in July.
-
Content Derivatives: Key methods from all three projects will serve as practical case studies for the ARA evaluation series, scheduled for simultaneous release across Bilibili, Zhihu, and GitHub.