Sulfonylhydrazones as novel antimalarial scaffold: design, synthesis, biological evaluation and machine learning models for structure-activity relationship interpretation

Vol. 1, 2019 - 111974
Poster and Oral (selected)
Favorite this paper
How to cite this paper?
Abstract

Among the top 10 causes of death in low-income countries, malaria not only presents high mortality but also high morbidity (219 million people).1 Although artemisinin-based combination therapy (ACT) remains effective, discovering new drugs is crucial to avoid resistance upsurge and ACT discard.1 To obtain novel hits against malaria with low toxicity, sulfonyl chlorides, hydrazine hydrate and aromatic aldehydes/ketones were utilized to produce 35 sulfonylhydrazones. Biological evaluation was performed against P. falciparum chloroquine-resistant (W2) and sensitive (3D7) strains and cellular viability employing WI-26VA-4 strain of human cells. 22 compounds were more potent than chloroquine against W2 strain, 18 IC50 values were nanomolar against 3D7 strain and cytotoxicity values were in the chloroquine’s logarithmic unit. To establish structure-activity relationship, decision tree machine learning predictor models were constructed and capable to classify active/inactive compounds with accuracy2 of 91.4% for 3D7 and 82.9% for W2 employing fingerprints and 94.3% and 91.4% using molecular descriptors.

Institutions
  • 1 University of São Paulo/School of Pharmaceutical Sciences
  • 2 Biosciences Institute/Federal University of São João del-Rei
Track
  • 1. Strategies in Drug Design
Keywords
malária
Machine Learning
decision tree
sulfonylhydrazone
LBDD