Drug discovery
Pharma · Biotech · CROs
Find intervention opportunities in disease trajectories. Prioritize targets, drugs and combinations, then connect candidates with experimental validation.
Disease models / Top lists / Candidate packages
01 / THE PLATFORM
A continuous research workflow connects data, models, candidates and validation results. Each project contributes reusable knowledge and model assets.
High-resolution representations × disease dynamics × mechanisms
Validation feeds the next model iterationSingle-cell, multi-omics, spatial transcriptomics, pathology images and clinical outcomes.
scRNA / scATAC / Spatial / ClinicalHigh-resolution cell representations, cross-modal alignment, cohort expansion and disease trajectories.
DOLPHIN / MATES / scCross / scGALA / scSemiProfilerSimulate drug and gene interventions to rank candidates, identify responsive cell types and explain key mechanisms.
UNAGI / CellAgentChat / DENetwork / SIDISH / veloAgentConnect with cell, organoid, organ-on-chip and human tissue studies to deliver reports and candidate packages.
Top List / Candidate Package / SaaS / API02 / FOUR APPLICATIONS
The disease, cohort or efficacy question determines the models, deliverables and validation approach.
Pharma · Biotech · CROs
Find intervention opportunities in disease trajectories. Prioritize targets, drugs and combinations, then connect candidates with experimental validation.
Disease models / Top lists / Candidate packagesHospitals · Clinical research teams
Bring together single-cell data, multiple modalities and clinical outcomes for patient stratification, high-risk cell analysis and reviewable agent-generated research reports.
Cohort analysis / SIDISH Agent / On-premises deploymentHealth management · Aging research
Build individual and population cell-state models, prioritize intervention ingredients and refine assessments through longitudinal profiling and follow-up.
State models / Ingredient prioritization / Longitudinal follow-upAesthetic clinics · Skincare brands · CROs
Connect ingredient screening, mechanism analysis and efficacy validation across skin stratification, barrier repair, inflammation and aging.
Skin stratification / Ingredient ranking / Efficacy validation03 / RESEARCH & VALIDATION
From human tissue validation to hospital research workflows, computational results connect with biological experiments in a shared evidence chain.
Nature Biomedical Engineering · 2025UNAGI models idiopathic pulmonary fibrosis (IPF) progression, prioritizes drug candidates and compares predictions with treatment responses in human precision-cut lung slices (PCLS).
Natural-language research requests connect to SIDISH tools to generate four report sections for clinician review: patient/cohort profiles, high-risk cell characteristics, virtual perturbations and integrated findings.
Work with Luohua Biology and other validation partners on cell, organoid and organ-on-chip studies. Candidate packages bring together rankings, mechanism evidence, validation results and development recommendations.
04 / THE SCIENCE
Published research from Ding Lab spans cell representations, cohort expansion, disease dynamics, communication networks and target discovery.

Track cell states through disease progression and prioritize drug candidates with the potential to reverse disease states.

Combine deep generative models and active learning to infer cohort-wide single-cell profiles from measured representative samples.

Integrate single-cell and bulk transcriptomics to identify high-risk cells, analyze survival and prioritize candidate targets.
05 / THE TEAM
Scientific strategy, project delivery and clinical statistics work together, supported by a team developing, evaluating and delivering models and platforms.

Scientific strategy · Platform direction · International research
Associate Professor in the Faculty of Medicine at McGill University and an affiliate member of Mila. Research in computational biology, single-cell omics and AI disease modeling.
Academic profile
Project management · Industry research · Delivery
Coordinates project scope, schedules, acceptance and product delivery, with cross-functional collaboration and bilingual communication.

Biostatistics · Cohort design · Translational validation
Leads endpoint definition, statistical methods, clinical cohort studies and real-world data analysis, connecting data quality with interpretation.

Single-cell modeling & active learning

Single-cell deep learning

Multimodal biomedical AI

Disease dynamics & drug discovery

Single-cell omics & AI drug discovery
06 / WORK WITH US
Choose the depth of collaboration around your research needs, from project analysis and validation to licensing and co-development.
Address a defined disease or efficacy question with models, candidate lists, mechanism reports and validation plans.
Bring candidate priorities, mechanistic evidence and experimental results together for the next development decision.
Support ongoing analysis, automated reporting and versioned model assets through SaaS, APIs or on-premises deployment.
Advance candidates, indications and efficacy-focused products with industry partners through milestone-based collaboration.
PARTNER WITH MIMESIS
Tell us about your disease area, data and research goals. Together, we can define the models, deliverables and validation pathway.
Contact the team jun.ding@mcgill.ca