Recognizing and validating ligands with CheckMyBlob

Abstract

Structure-guided drug design depends on the correct identification of ligands in crystal structures of protein complexes. However, the interpretation of the electron density maps is challenging and often burdened with confirmation bias. Ligand identification can be aided by automatic methods such as CheckMyBlob, a machine learning algorithm that learns to generalize ligand descriptions from sets of moieties deposited in the Protein Data Bank. Here, we present the CheckMyBlob web server, a platform that can identify ligands in unmodeled fragments of electron density maps or validate ligands in existing models. The server processes PDB/mmCIF and MTZ files and returns a ranking of 10 most likely ligands for each detected electron density blob along with interactive 3D visualizations. Additionally, for each prediction/validation, a plugin script is generated that enables users to conduct a detailed analysis of the server results in Coot. The CheckMyBlob web server is available at https://checkmyblob.bioreproducibility.org.

Graphical Abstract

CheckMyBlob is a machine learning system that automatically detects and validates ligands in X-ray electron density maps.

Figure 1.

Schematic of the ligand clustering procedure. Arrows depict example cluster splits based on different criteria at subsequent stages of the process.

Figure 2.

Schematic representation of the CheckMyBlob workflow and screenshots of the interactive results visualization page and Coot ligand analysis script. The user provides input files (an MTZ file and PDB or mmCIF file) and chooses to either detect unmodeled ligands or validate existing ligands. Next, blobs are detected, extracted from electron density maps, and described by a set of numerical features. The obtained numerical features are input to a machine learning model, which outputs a ranking of the ten most likely ligands for each blob. This probability-based ranking can be viewed on the interactive results visualization page and tested in Coot through a downloadable script.