The leishmaniases are a group of neglected tropical diseases (NTDs) endemic in Tunisia, and more broadly in the Middle East and North Africa. While novel treatments against this group of diseases is of utmost importance to the country, the private sector has taken little to no interest in them because they do not represent a lucrative market and end-to-end drug discovery processes are known to be long and costly. Innovative and cost-effective approaches are needed, including computational approaches and drug repurposing. This project aimed to develop cutting-edge computational approaches coupled to biological assays towards drug discovery and repurposing against leishmaniasis. The methodology can also be applied to other diseases, and with the advent of the COVID-19 pandemic, the team was able to rapidly and successfully adapt their approaches and technologies and participate in the universal research efforts to tackle COVID-19, through developing accurate algorithms for anti-coronavirus molecules.

Final Summary of Project Activities

Dr. Harigua and her fellow researchers developed datasets on molecules presenting anti-pathogenic activity for leishmania and SARS-COV-2, through literature review and database searches for ligand-based drug discovery. They considered multiple criteria in order to consider these datasets for artificial intelligence (AI) applications. The team implemented seven machine learning (ML) and four deep learning (DL) algorithms on both datasets. The trained algorithms were compared and those that performed best were used for modeling the potential repurposing of existing FDA-approved drugs. The team then set up infection models for drug discovery and identified molecules from FDA-approved drugs that were predicted as potentially active. Those drugs were then tested for their effects on the growth of promastigotes, an extracellular form of the leishmania parasite.

The researchers have published two papers on their work and have others forthcoming from these results. They also presented their findings at conferences of the American Society of Tropical Medicine and Hygiene and the African Society of Bioinformatics and Computational Biology.

During the project, two undergraduate students from an engineering school were trained for six months each to complete their degree in statistics and data analysis. Dr. Harigua also mentored two scientists, including one who completed and defended her doctoral work and one who submitted her first research proposal as a PI. Dr. Harigua also received an additional $83,000 grant alongside one of her mentees and others at the Institut Pasteur de Tunis for future work.

Publications

Emna Harigua-Souiai, Rafeh Oualha, Oussama Souiai, Ines Abdeljaoued-Tej, and Ikram Guizani. 2022. Applied machine learning toward drug discovery enhancement: leishmaniases as a case study. Bioinformatics and Biology Insights 16: 1-10. https://doi.org/10.1177/11779322221090349

Emna Harigua-Souiai, Mohamed Mahmoud Heinhane, Yosser Zina Abdelkrim, Oussama Souiai, Ines Abdeljaoued-Tej, and Ikram Guizani. 2021. Deep learning algorithms achieved satisfactory predictions when trained on a novel collection of anticoronavirus molecules. Frontiers in Genetics 12-2021. https://doi.org/10.3389/fgene.2021.744170