Multitask Learning-Driven Identification of Novel Antitrypanosomal Compounds

Abstract

Background: Chagas disease and human African trypanosomiasis cause substantial death and morbidity, particularly in low- and middle-income countries, making the need for novel drugs urgent. Methodology & results: Therefore, an explainable multitask pipeline to profile the activity of compounds against three trypanosomes (Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense and Trypanosoma cruzi) were created. These models successfully discovered four new experimental hits (LC-3, LC-4, LC-6 and LC-15). Among them, LC-6 showed promising results, with IC₅₀ values ranging 0.01–0.072 μM and selectivity indices >10,000. Conclusion: These results demonstrate that the multitask protocol offers predictivity and interpretability in the virtual screening of new antitrypanosomal compounds and has the potential to improve hit rates in Chagas and human African trypanosomiasis projects.

Tweetable abstract

Improved approaches to antitrypanosomal drug discovery are needed to boost hit rates. To address this issue, multitask models were created to profile the activity of compounds against three trypanosomes. These models facilitated the discovery of four new antitrypanosomal hits.

Graphical Abstract

Keywords:

• deep learning
• low-data regimes
• model explainability
• neglected tropical diseases
• QSAR
• trypanosomatids
• virtual screening

Supplementary data

To view the supplementary data that accompany this paper please visit the journal website at:www.tandfonline.com/doi/full/10.2217/epi-2016-0184

Author contributions

The manuscript was written with contributions from all authors. All authors gave approval to the final version of the manuscript.

Acknowledgments

The authors would like to thank Brazilian funding agencies, the National Counsel of Technological and Scientific Development (CNPq), the Coordination of Improvement of Higher Education Personnel and the French Committee for the Evaluation of University Cooperation with Brazil (CAPES/COFECUB) program, State of Goiás Research Foundation (FAPEG) for financial support and fellowships.

Financial & competing interests disclosure

This work was funded by CNPq (425119/2018-1), FAPEG (202110267000293), CAPES (Finance Code 001) and the CAPES/COFECUB program (88881.711954/2022-01). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. RC Braga is CTO of InsilicAll Inc. EN Muratov is cofounder of Predictive, LLC, which develops computational methodologies and software for toxicity prediction. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.