Ersilia Book
  • 🤗Welcome to Ersilia!
    • The Ersilia Open Source Initiative
    • Ten principles
    • Ersilia's ecosystem
  • 🚀Ersilia Model Hub
    • Getting started
    • Online inference
    • Local inference
    • Model contribution
      • Model template
      • Model incorporation workflow
      • Troubleshooting models
      • BioModels annotation
    • For developers
      • Command line interface
      • CI/CD workflows
      • Test command
      • Testing playground
      • Model packaging
      • Inputs
      • Codebase quality and consistency
      • Results caching
  • 💊Chemistry tools
    • Automated activity prediction models
      • Light-weight AutoML with LazyQSAR
      • Accurate AutoML with ZairaChem
      • Model distillation with Olinda
    • Sampling the chemical space
    • Encryption of AI/ML models
  • AMR chemical collections
  • 🙌Contributors
    • Communication channels
    • Tech stack
    • Internships
      • Outreachy Summer 2025
      • Outreachy Winter 2024
      • Outreachy Summer 2024
      • Outreachy Winter 2023
      • Outreachy Summer 2023
      • Outreachy Winter 2022
      • Outreachy Summer 2022
  • 📑Training materials
    • AI2050 intro workshop
    • AI2050 AI for Drug Discovery
    • Introduction to ML for Drug Discovery
    • Python 101
    • External resources
  • 🎨Styles
    • Brand guidelines
    • Slide and document templates
    • Scientific figures with Stylia
    • Coding style
  • 🌍About Us
    • Where to find us?
    • Diversity and inclusion statement
    • Code of conduct
    • Open standards and best practices
    • Ersilia privacy notice
    • Strategic Plan 2025-2027
    • Ersilia, the Invisible City
Powered by GitBook

2025, Ersilia Open Source Initiative

On this page
  • 1. Select your model of interest
  • 2. Prepare your input data
  • 3. Run predictions and download results
  • 4. Check your predictions

Was this helpful?

  1. Ersilia Model Hub

Online inference

Documentation to run Ersilia models online

PreviousGetting startedNextLocal inference

Last updated 1 month ago

Was this helpful?

To ensure users from all backgrounds are able to benefit from our tools, we provide a ready-to-use, no-code solution to obtain predictions for your molecules.

1. Select your model of interest

We offer a broad range of models, from bioactivity prediction against several pathogens (malaria, tuberculosis, schistosomiasis, ESKAPE pathogens...) to ADME endpoints and toxicity predictions. Use our to browse models according to your needs and take note of the model identifier you wish to use!

Please ensure you understand the output of each model before using it. For example, a classifier will output the probability of being active in a particular assay, e.g. probability that a molecule kills the malaria parasite in an in vitro assay, and a regressor might output the predicted IC50 value at which the molecule inhibits the growth of the parasite.

2. Prepare your input data

The molecules must be displayed in SMILES notation. You can use to find the SMILES notation of a given compound: simply introduce the compound name on the search bar (for example, aspirin), select the best result and scroll to the SMILES section within "Name and Identifiers" (in this case; CC(=O)OC1=CC=CC=C1C(=O)O). If your starting input data is an .sdf file, use your preferred visualiser, like ChemDraw, to open the molecule and obtain its SMILES representation. To deal with multiple molecular file formats, including SDF, you can use to convert them into SMILES notation. Alternatively, you can also use free software like to draw a molecule and then simply click on save 💾 it as a SMILES.

Collect your list of SMILES in a .csv or .txt file.

3. Run predictions and download results

Go to our and select your model of choice from the drop down list. Copy the list of SMILES (maximum allowed 100 molecules) and click on "Run Predictions!". Wait a few minutes to download your results!

If you wish to run larger annotations, for example running several predictions against a database of >1k molecules, please contact Ersilia directly to obtain a customised solution: hello@ersilia.io

4. Check your predictions

By default, Ersilia will provide a downloadable .csv file summarizing the results, containing the following columns:

  • SMILES: the input SMILES (please note that these might have been standardised if they were not provided in the standard format).

  • InChIKey: 27-character unique identifier of the molecule based on the International Chemical Identifier (InChI).

  • Model output: one or several columns containing the predictions of the selected model. Make sure to read about the model in the literature or in the Ersilia documentation to appropriately interpret the model's results.

Posting to this free online service will make your molecules public. Please consider if you are working with IP-protected molecules.

🚀
dynamic interface
PubChem
OpenBabel
Marvin.js
online inference app
local inference