Cytotoxicity Prediction

Cytotoxicity in silico

The Way2Drug platform offers researchers a unique set of tools for predicting the cytotoxicity of compounds against human cell lines, eliminating the need for costly and time-consuming experimental studies.

Platform Services

In late 2025, CLC-Pred 2.0 was further extended with QSAR models for the quantitative prediction of pIC₅₀ and pGI₅₀ values for 10 normal and 10 tumor cell lines most commonly used in toxicological studies.

Applications

The cytotoxicity prediction services available on the Way2Drug platform cover a broad spectrum of research tasks in drug discovery and preclinical safety assessment. Early toxicity identification is one of the primary applications. By predicting cytotoxic activity against a diverse panel of normal human cell lines, researchers can evaluate organ-specific toxicity profiles at the earliest stages of compound development. Hepatotoxicity can be assessed using the HepG2 liver cell line, nephrotoxicity using HEK-293 and HEK-293T kidney cells, immunotoxicity using peripheral blood mononuclear cells, pulmonary toxicity using MRC-5 lung fibroblasts, and vascular toxicity using HUVEC endothelial cells. This multi-organ coverage allows for a comprehensive picture of the potential toxic burden of a compound long before any animal experiments are conducted.

Selection of promising drug candidates is greatly facilitated by the availability of quantitative predictions. The IC₅₀ and GI₅₀ values predicted for both tumor and normal cell lines enable the calculation of the Therapeutic Index, a key parameter reflecting the ratio of a compound's toxic to therapeutically effective concentrations. Evaluating the therapeutic window at the in silico stage helps prioritize only those candidates that demonstrate a meaningful selectivity margin between tumor and normal cells, thereby reducing the risk of late-stage attrition in drug development.

Lead compound optimization benefits from the simultaneous prediction of cytotoxic activity across a panel of cell lines representing different tissues and tumor types. By comparing predicted activity profiles across multiple cell lines in a single query, medicinal chemists and pharmacologists can identify structural features associated with undesirable toxicity toward normal tissues or insufficient activity against target tumor cell lines, and use this information to guide the rational modification of chemical structures.

Drug repositioning represents another valuable application, as the broad coverage of cell lines and molecular mechanisms of action in CLC-Pred 2.0 allows the identification of new potential indications for already approved drugs or compounds previously studied in other therapeutic contexts. Combined with the prediction of over 2 170 molecular mechanisms of action, the platform enables hypothesis-driven repositioning strategies based on both phenotypic and mechanistic evidence.

Finally, the tools are well suited for the investigation of natural compounds and plant-derived extracts. Since many natural products have complex, polypharmacological profiles and are tested against diverse biological targets, the ability to rapidly screen their cytotoxic potential against hundreds of cell lines in silico allows researchers to prioritize the most promising substances for subsequent experimental validation and to better understand their mechanisms of action.

Scientific Validation

The accuracy of the qualitative SAR models in CLC-Pred 2.0 is confirmed by a mean AUC of 0.925 under LOO cross-validation; for the quantitative QSAR models, the mean R² 5-fold CV and RMSE values are 0.691 and 0.584, respectively. Since the launch of the original CLC-Pred, more than 80 peer-reviewed publications have employed these tools to evaluate the cytotoxicity of novel synthetic and natural compounds.