ACD/Percepta physicochemical and adme tox prediction software

ACD/P450 Regioselectivity and Metabolism

Predict metabolic soft spots for metabolism by human liver microsomes and Cytochrome P450 enzymes from structure

The CYP-P450 Regioselectivity module in ACD/Percepta provides valuable insights into a compound's metabolic profile early in the drug discovery process when little or no experimental information is available, and labor intensive investigation of each compound in the screening process is prohibited by the large number of compounds involved. The prediction module can help identify metabolic sites on new chemical entities; guide synthesis of compounds with improved metabolic properties; and help identify and elucidate likely metabolite structures.

ACD/Regioselectivity of Metabolism: CYP1A2 regioselectivity
Click to view the image series
  • Predict metabolic soft spots for metabolism by:
    • Human liver microsomes (HLM)
    • Five major Cytochrome P450 enzymes (CYP3A4, CYP2D6, CYP2C9, CYP2C19, and CYP1A2)
  • Interpret results quickly and easily with at-a-glance color-mapping on the structure
  • View metabolic reactions for each isoform—ranked from most probable, to least likely to occur
  • Browse analogous sites of metabolic liability in the internal training set
  • Judge the accuracy of prediction using the Reliability Index (RI)

Module Features

  • Predictions are based on probabilistic models, developed using GALAS (Global, Adjusted Locally According to Similarity) modeling methodology, for 5 most common metabolic reactions (N-dealkylation, O-dealkylation, aliphatic hydroxylation, aromatic hydroxylation, S-oxidation).
  • Predictions are colored-coded according to calculated metabolism scores (red—likely site of metabolism, score >0.6; green—confident prediction of the non-metabolized atom, score <0.4; grey—inconclusive predictions, score between 0.4 and 0.6) to easily visualize potentially active sites.
  • A quantitative estimate of prediction reliability is provided by Reliability Index (RI)—a number from 0 to 1 (0—poor reliability, 1—excellent reliability), indicating similarity of the current metabolic site to our internal training set, and consistency of model predictions for similar sites.

About the Model

Data source
Peer-reviewed original articles with analytical identification of the metabolites observed after the incubation of compound with human liver microsomes or recombinant cytochrome P450 enzymes. In these compounds each carbon atom, with at least one hydrogen attached, produced a new data point that was considered positive if a metabolic reaction taking place at that site was experimentally observed, and negative otherwise).

Training set size: >900 compounds

Isoform No. of Data Points
CYP 1A2 6464
CYP 2C19 5474
CYP 2C9 6645
CYP 2D6 6708
CYP 3A4 7728
Overall HLM data 8791

To complete the P450 prediction package, ACD/Percepta also offers modules for the prediction of substrates and inhibitors of the five major isoforms of Cytochrome P450, as a complement to early high-throughput screening assays. These modules are trainable with in-house data, so that models better reflect your chemical space.

ACD/Labs Product Suites

The Percepta prediction modules are available as bundles to offer cost savings for multiple modules, and provide related modules as a package. Contact us for more information on the product suite that is right for you.