Ginkgo Datapoints launched the Virtual Cell Pharmacology Initiative (VCPI) to generate standardized pharmacology datasets for virtual cell modeling. The program will offer a reference engineered cell line (V-Ref293), free high-throughput RNA profiling to contributors, and intends to test at least 100,000 compounds, producing over 12 billion data points for model training. John Androsavich, PhD, general manager of Ginkgo Datapoints, framed VCPI as an effort to align wet‑lab standards with AI training needs; the Arc Institute’s Virtual Cell Challenge will run in parallel. The initiative addresses persistent reproducibility and data-quality gaps that have limited the predictive power of virtual cell models for drug discovery.