Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase.
Thesis (PhD)--Stellenbosch University, 2025.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | en_ZA |
| Published: |
Stellenbosch : Stellenbosch University
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613779585400832 |
|---|---|
| access_status_str | Open Access |
| author | Thibaud, Jessica Leigh |
| author2 | De Villiers, Katherine A. |
| author_browse | De Villiers, Katherine A. Thibaud, Jessica Leigh |
| author_facet | De Villiers, Katherine A. Thibaud, Jessica Leigh |
| author_sort | Thibaud, Jessica Leigh |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (PhD)--Stellenbosch University, 2025. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/134392 |
| institution | Stellenbosch University (South Africa) |
| language | en_ZA |
| last_indexed | 2026-06-10T12:41:34.416Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/134392 Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase. Thibaud, Jessica Leigh De Villiers, Katherine A. Stellenbosch University. Faculty of Science. Dept. of Chemistry & Polymer Science. Malaria -- Prevention Machine learning Protein kinases -- Inhibitors Plasmodium falciparum UCTD Thesis (PhD)--Stellenbosch University, 2025. Thibaud, J. L. 2025. Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase. Unpublished doctoral thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/1ea45598-90e3-4481-a4b8-7dda32709fbd ENGLISH ABSTRACT: The identification of antiplasmodium treatments with new mechanisms of action able to circumvent parasite resistance is crucial. While traditional high-throughput screening is time-consuming and resource intensive, the advancement of in silico methods has substantially expanded the drug-discovery toolbox for the purpose of identifying potent, selective and well-tolerated drugs. The work presented in this dissertation describes the development of an in silico workflow consisting of various ligand- and structure-based approaches for the rational identification of new inhibitors of Plasmodium falciparum cGMP-dependent protein kinase (PfPKG) and hemozoin formation. Both PfPKG and hemozoin formation are essential to parasite survival, allowing for merozoite egress and heme detoxification, respectively, during the parasite asexual blood stage (ABS). Within the workflow, two ligand-based approaches were applied. Firstly, principal component analysis (PCA) was used to investigate the tranche of two-dimensional antiplasmodium chemical space occupied by the Tres Cantos Antimalarial Dataset (TAMS). The projection and visualisation of known PfPKG and β-hematin (synthetic hemozoin) inhibitors onto this space resulted in regions of inhibitor enrichment that could be used in a predictive capacity to a test set of compounds with unknown activity. Secondly, a Random Forest (RF) Machine Learning (ML) classification model was developed to refine predictions of PfPKG inhibition. In addition, molecular-modelling was employed as a structure-based approach to investigate ligand-receptor interactions between test compounds and crystal structures of Plasmodium PKG and β-hematin. Initially, the workflow was applied to a number of structurally-diverse ChemDiv libraries for the prediction of PfPKG inhibition. This resulted in the identification of potent new PfPKG inhibitor, D8 (IC50 = 88.0 ± 3.0 nM), which was recognised as being the FDA-approved anticancer small-molecule kinase inhibitor, ibrutinib. Subsequently, a refined version of the in silico workflow was applied to a scaffold-focused library of monocyclic trisubstituted azole-based compounds. This was motivated by recent studies in the literature describing the identification of trisubstituted imidazoles and thiazoles with potent activity against PfPKG. In this regard, however, the workflow was largely unsuccessful with only eight of the 50 purchased predicted actives obtaining ≥ 25.0% enzyme inhibition. The most active compound achieved an average PfPKG percentage inhibition of 51.6% and IC50 value of 3.8 ± 0.1 μM which is 195× less active than the positive control (MLN0128). An interrogation of possible limitations indicated that the model is unable to accurately predict compound potency. Thus, findings suggest that the application of a local model may be better suited for the investigation of a scaffold-focused library specific to trisubstituted azoles. Although predominantly unsuccessful when used to investigate a scaffold-focused library, the workflow achieved its aim of PfPKG inhibitor identification when applied to a set of structurally diverse ChemDiv datasets. Therefore, the refined workflow was applied to a structurally-diverse, target-focused library of FDA-approved small-molecule kinase inhibitors (‘ib’ compounds). More specifically, the final study describes the investigation of the dual-target activity of ‘ib’ compounds against PfPKG and β-hematin formation, applying the concepts of drug-repurposing and polypharmacology to help combat antiplasmodium drug-resistance. Of the 12 purchased ‘ib’ compounds, five showed dual activity against β-hematin formation and the PfPKG enzyme to some extent, while two were identified with sub-micromolar PfPKG activity and IC50 values less than 100.0 μM to β-hematin. These include derazantinib and ibrutinib, both FDA-approved treatments of various cancers. Importantly, derazantinib, which also shows potent whole cell activity against the chloroquine-sensitive NF54 cell line, contains the benzoquinazoline scaffold which has not previously been evaluated against PfPKG, making it an interesting scaffold for future polypharmacological efforts in the fight against malaria. AFRIKAANSE OPSOMMING: Geen opsomming beskikbaar. Doctoral 2025-11-19T07:26:06Z 2025-11-19T07:26:06Z 2025-03 Thesis https://scholar.sun.ac.za/handle/10019.1/134392 en_ZA Stellenbosch University xvi, 308 pages : illustrations, includes annexures application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Malaria -- Prevention Machine learning Protein kinases -- Inhibitors Plasmodium falciparum UCTD Thibaud, Jessica Leigh Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase. |
| title | Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase. |
| title_full | Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase. |
| title_fullStr | Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase. |
| title_full_unstemmed | Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase. |
| title_short | Machine learning approaches towards identifying inhibitors for plasmodium falciparum cGMP-dependent protein kinase. |
| title_sort | machine learning approaches towards identifying inhibitors for plasmodium falciparum cgmp dependent protein kinase |
| topic | Malaria -- Prevention Machine learning Protein kinases -- Inhibitors Plasmodium falciparum UCTD |
| url | https://scholar.sun.ac.za/handle/10019.1/134392 |
| work_keys_str_mv | AT thibaudjessicaleigh machinelearningapproachestowardsidentifyinginhibitorsforplasmodiumfalciparumcgmpdependentproteinkinase |