Edge Mapper for RBFE calculations [MDRun] [FECalc]

Category Paths

  • Product-based/Molecular Dynamics

  • Role-based/Computational Chemist

  • Role-based/Medicinal Chemist

  • Task-based/Data Science/Filtering

  • Solution-based/Hit to Lead/Affinity Prediction/Free-Energy Calculations


  • Purpose:

    • This Floe produces ligand pairs or edges to run relative binding free energy (RBFE) calculations

  • Method Recommendations/Requirements:

    • One inputs is required:

      • A ligand dataset. The provided ligands must have reasonable 3D coordinates, all atoms, and correct chemistry such as bond orders and formal charges should be correctly assigned

      • Optionally, the user can also import into the mapper Floe a map generated externally, to generate a mapper output dataset of only those edges defined by the external map. The syntax of this external map file is described below

  • Limitations

    • The provided ligand set should contain similar ligands where a common scaffold is used and functional groups are added to it. Large scaffold mutations should be avoided to improve RBFE and ABFE result errors

  • Expertise Level:

    • Regular

  • Compute Resource:

    • low

  • Keywords:

    • MD, FECalc

  • Related Floes: * Non-Equilibrium Switching Recovery [Utility] [FECalc]

In general given N ligands N(N-1)/2 edges can be used to run RBFE calculations. The mapper Floe selects a reasonable subset of these edges with the goal to keep the computational demand low. The edges are selected based on set of heuristics like ligand mcs similarities, equal charges and OpenEye rocs score. The produced Floe report shows the similarity scores as numbers in the range [0, 1] where zero means very unlikely and 1 means very likely to succeed. The Floe mapper output can be used with the NES Floe to run RBFE calculations. In order to have successful RBFE results the ligands should be similar and do not attempt large mutations. The Mapper Floe can accept a edge text input file as well. In this case the produced map is defined in the provide text file and no attempt to use internal heuristics is done. The text file of edges, describes the map of desired alchemical transformations of one ligand into another; each transformation forms an edges of a connected graph of ligands. The file must have one line per transformation, of format

ligA_name >> ligB_name

where “ligA_name” and “ligB_name” are the ligand title names for the ligands to be transformed. These ligand names must correspond exactly to those in the provided ligand set

Promoted Parameters

Title in user interface (promoted name)

  • Optional Ligand Edge Map file (edges) type: file_in: A User-defined ligand edge map text file
  • Ligand or Bound Equilibrium Dataset (ligands) type: data_source: Ligand or Bound Equilibrium Input Dataset
  • Output Dataset (mapper) type: dataset_out: Output dataset to write to
  • All Edges Mapper Dataset (out_all) type: dataset_out: All N*(N-1)/2 possible edges Output Dataset
    Default: all_possible_edges_dataset_output