Similarity Search of Flexible 3D Molecules Combining Local and Global Shape Descriptors

AI Promoted Virtual Screening, Structure-Based Hit Optimization, and Synthesis of Novel COVID-19 S-RBD Domain Inhibitors

Coronavirus disease 2019 (COVID-19) is caused by a new, highly pathogenic severe-acute-respiratory syndrome coronavirus 2 (SARS-CoV-2) that infects human cells through its transmembrane spike (S) glycoprotein. The receptor-binding domain (RBD) of the S protein interacts with the angiotensin-converting enzyme II (ACE2) receptor of the host cells. Therefore, pharmacological targeting of this interaction might prevent infection or spread of the virus. Here, we performed a virtual screening to identify small molecules that block S-ACE2 interaction. Large compound libraries were filtered for drug-like properties, promiscuity and protein-protein interaction-targeting ability based on their ADME-Tox descriptors and also to exclude pan-assay interfering compounds. A properly designed AI-based virtual screening pipeline was applied to the remaining compounds, comprising approximately 10% of the starting data sets, searching for molecules that could bind to the RBD of the S protein. All molecules were sorted according to their screening score, grouped based on their structure and postfiltered for possible interaction patterns with the ACE2 receptor, yielding 31 hits. These hit molecules were further tested for their inhibitory effect on Spike RBD/ACE2 (19-615) interaction. Six compounds inhibited the S-ACE2 interaction in a dose-dependent manner while two of them also prevented infection of human cells from a pseudotyped virus whose entry is mediated by the S protein of SARS-CoV-2. Of the two compounds, the benzimidazole derivative CKP-22 protected Vero E6 cells from infection with SARS-CoV-2, as well. Subsequent, hit-to-lead optimization of CKP-22 was effected through the synthesis of 29 new derivatives of which compound CKP-25 suppressed the Spike RBD/ACE2 (19-615) interaction, reduced the cytopathic effect of SARS-CoV-2 in Vero E6 cells (IC50 = 3.5 μM) and reduced the viral load in cell culture supernatants. Early in vitro ADME-Tox studies showed that CKP-25 does not possess biodegradation or liver metabolism issues, while isozyme-specific CYP450 experiments revealed that CKP-25 was a weak inhibitor of the CYP450 system. Moreover, CKP-25 does not elicit mutagenic effect on Escherichia coli WP2 uvrA strain. Thus, CKP-25 is considered a lead compound against COVID-19 infection.

DeepSurf: A surface-based deep learning approach for the prediction of ligand binding sites on proteins

MOTIVATION

The knowledge of potentially druggable binding sites on proteins is an important preliminary step towards the discovery of novel drugs. The computational prediction of such areas can be boosted by following the recent major advances in the deep learning field and by exploiting the increasing availability of proper data.

RESULTS

In this paper, a novel computational method for the prediction of potential binding sites is proposed, called DeepSurf. DeepSurf combines a surface-based representation, where a number of 3 D voxelized grids are placed on the protein's surface, with state-of-the-art deep learning architectures. After being trained on the large database of scPDB, DeepSurf demonstrates superior results on three diverse testing datasets, by surpassing all its main deep learning-based competitors, while attaining competitive performance to a set of traditional non-data-driven approaches.

AVAILABILITY

The source code of the method along with trained models are freely available at https://github.com/stemylonas/DeepSurf.git.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Three-Dimensional Krawtchouk Descriptors for Protein Local Surface Shape Comparison

Direct comparison of three-dimensional (3D) objects is computationally expensive due to the need for translation, rotation, and scaling of the objects to evaluate their similarity. In applications of 3D object comparison, often identifying specific local regions of objects is of particular interest. We have recently developed a set of 2D moment invariants based on discrete orthogonal Krawtchouk polynomials for comparison of local image patches. In this work, we extend them to 3D and construct 3D Krawtchouk descriptors (3DKDs) that are invariant under translation, rotation, and scaling. The new descriptors have the ability to extract local features of a 3D surface from any region-of-interest. This property enables comparison of two arbitrary local surface regions from different 3D objects. We present the new formulation of 3DKDs and apply it to the local shape comparison of protein surfaces in order to predict ligand molecules that bind to query proteins.

Automated shape-based clustering of 3D immunoglobulin protein structures in chronic lymphocytic leukemia

Background

Although the etiology of chronic lymphocytic leukemia (CLL), the most common type of adult leukemia, is still unclear, strong evidence implicates antigen involvement in disease ontogeny and evolution. Primary and 3D structure analysis has been utilised in order to discover indications of antigenic pressure. The latter has been mostly based on the 3D models of the clonotypic B cell receptor immunoglobulin (BcR IG) amino acid sequences. Therefore, their accuracy is directly dependent on the quality of the model construction algorithms and the specific methods used to compare the ensuing models. Thus far, reliable and robust methods that can group the IG 3D models based on their structural characteristics are missing.

Results

Here we propose a novel method for clustering a set of proteins based on their 3D structure focusing on 3D structures of BcR IG from a large series of patients with CLL. The method combines techniques from the areas of bioinformatics, 3D object recognition and machine learning. The clustering procedure is based on the extraction of 3D descriptors, encoding various properties of the local and global geometrical structure of the proteins. The descriptors are extracted from aligned pairs of proteins. A combination of individual 3D descriptors is also used as an additional method. The comparison of the automatically generated clusters to manual annotation by experts shows an increased accuracy when using the 3D descriptors compared to plain bioinformatics-based comparison. The accuracy is increased even more when using the combination of 3D descriptors.

Conclusions

The experimental results verify that the use of 3D descriptors commonly used for 3D object recognition can be effectively applied to distinguishing structural differences of proteins. The proposed approach can be applied to provide hints for the existence of structural groups in a large set of unannotated BcR IG protein files in both CLL and, by logical extension, other contexts where it is relevant to characterize BcR IG structural similarity. The method does not present any limitations in application and can be extended to other types of proteins.