Streamline Your Cheminformatics: Dimorphite-DL is Now Available on Vecura
This update enables medicinal chemists and computational scientists to accurately enumerate molecular ionization states through a guided workflow inside Vecura, without setting up complex technical infrastructure.
What is Dimorphite-DL?
Dimorphite-DL is a fast, open-source cheminformatics tool designed to enumerate the ionization (protonation) state variants of drug-like small molecules across a specified pH range. By using a rule-based SMARTS pattern matching approach, it identifies ionizable functional groups and calculates their predominant protonation states without the need for computationally heavy neural networks.
It helps users ensure their molecular structures are accurately represented before downstream tasks. It is especially useful for drug discovery workflows, including molecular docking, QSAR model training, and any property-prediction pipeline where precise hydrogen counts are critical for accuracy.
What can users do with Dimorphite-DL on Vecura?
With Dimorphite-DL on Vecura, users can:
- Automatically generate all plausible protonation microspecies for a given molecule within a defined pH window.
- Customize the pH range and precision settings to match specific physiological or experimental environments.
- Easily integrate protonation state enumeration into automated drug-discovery pipelines.
- Generate visual representations of these variants to better understand the chemical behavior of their compounds.
What the output means
The output provides a set of canonical SMILES strings representing each enumerated ionization microspecies, along with optional state labels and visual structural previews.
This output should be used to support scientific decision making. It does not replace experimental validation.
Why this matters
In drug discovery, the biological activity and binding affinity of a molecule are heavily influenced by its protonation state at physiological pH. Incorrectly modeling these states can lead to poor docking results or inaccurate predictions in machine learning models, effectively derailing research efforts before they gain momentum.
By automating this process, Dimorphite-DL removes a significant bottleneck in molecular preprocessing. It ensures that researchers are working with the most likely chemical species present in the body, leading to more reliable, reproducible, and scientifically robust data in the early stages of drug development.
- Developed by: The Durrant Lab
- Source: GitHub, PyPI, and Journal of Cheminformatics
- Reference: Ropp et al., J Cheminform 2019
Try Dimorphite-DL on Vecura.
Open the model workspace and start evaluating it with your own inputs.