Skip to content
This repository was archived by the owner on Feb 11, 2026. It is now read-only.
This repository was archived by the owner on Feb 11, 2026. It is now read-only.

Multi-step retrosynthesis - input file format inquiry #49

Description

@higginss1

Good afternoon,

I've successfully been able to install and operate chemformer from the GH repo using the scripts in the provided example_scripts directory (e.g., predict.sh), but I'm unable to find documentation about the format of the input file for performing multi-step retrosynthesis of a given target product molecule with chemformer.

For example, if I want to perform multi-step retrosynthesis of the following target product (Nc1nc2[nH]c(CCCc3csc(C(=O)O)c3)cc2c(=O)[nH]1), can you provide an example of what the format of the input text file would need to be to get the molbart.retrosynthesis.round_trip_inference to work?

I have currently been using something like:

reactant product reaction_type set
NA Nc1nc2[nH]c(CCCc3csc(C(=O)O)c3)cc2c(=O)[nH]1 <RX_6> test

Note, I added NA above for reactant and a random value for reaction_type since I'm unsure as to what values are required there when multi-step retrosynthesis from a target product is desired. I may have missed it, but I've been unable to find documentation as to what to include in the input file when predicting reaction molecules for a given target product.

Any assistance you can provide or links to additional documentation on how to properly format the input file for multi-step retrosynthesis would be much appreciated.

Thank you and I'm looking forward to trying out chemformer for my multi-step retrosynthesis needs.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions