LeishTargets

Project looking for drug targets in the proteome of Leishmania species.

About

Leishmaniasis is a group of neglected infectious diseases, with approximately 1.3 million new cases each year, for which the available therapies have serious limitations due to factors including high cost, low efficacy, high toxicity and development of resistance on the part of parasites. Therefore, it is extremely important to apply efficient and low-cost methods capable of selecting the best therapeutic targets in order to speed up the development of new therapies against those diseases. However, most studies in drug discovery are based only on the analysis of a small target group selected in an ad hoc manner, neglecting new potentially druggable targets with therapeutic importance. Thus, in the present study, we integrated computational methods that evaluate proteins in the structural, chemical and functional context, and are able to infer the druggability of the predicted proteomes of Leishmania braziliensis and Leishmania infantum, species responsible for the different clinical manifestations of leishmaniasis in Brazil. In this way, our analyzes identified a group of pharmacologically attractive proteins and our methodology also allowed the identification of drugs already developed with probable ability to interact with these potential targets, thus providing for the possibility of reusing these compounds. Our results can be used by researchers in order to test the interaction of the group of compounds and proteins, thus allowing the reuse of these drugs and expanding the possibilities of treatment for Leishmaniasis. Our results also aid in the understanding of the Leishmania species druggable proteome and provide data on subcellular localization, molecular function, protein interaction networks and protein binding sites, previously absent in databases.

- Crhisllane Vasconcelos and Antonio Rezende

Paper 1

Building protein-protein interaction networks for Leishmania species through protein structural information.

Downloads Paper 1

tab1

Database SQL

All data in sql format.

Network_LinJ_paper1.tab

L. infantum

Predicted protein interaction network using structural information.

Network_LbrM_paper1.tab

L. braziliensis

Predicted protein interaction network using structural information.

gbm.fit_model.rds

GBM model

The machine learning algorithms were evaluated against positive and negative interaction datasets used as controls. Based on this analysis, the gbm technique showed a better performance when compared to other machine learning algorithms, obtaining an AUC = 0.88. The gbm algorithm calculates a response value ranging from 0 to 1, for which a minimum threshold of 0.46 has been determined based on controls to indicate interaction between the proteins.

Paper 2

Systematic in silico evaluation of Leishmania ssp. Proteomes for drug discovery.

Downloads Paper 2

Database

All data in sql format.

Binding pocket drug Similarity

Similarity of binding pocket drug by Methodology resume.

Druggability Similarity

Druggability Similarity methodology

Subcellular location Similarity

Subcellular location methodology.

Molecular function similarity

Molecular function similarity methodology.

Biological proccess similarity

Biological proccess similarity methodology.

Protein and Drug Target similarity

Protein and Drug Target similarity methodology.

Contact

Email
crhisllane.vasconcelos@ufpe.br