Inferring Strain Mixture within Clinical Plasmodium falciparum Isolates from Genomic Sequence Data

Area of Science:

• Genomics
• Parasitology
• Statistical Modeling

Background:

• Understanding Plasmodium falciparum (P. falciparum) population structure is crucial for malaria control.
• Mixed infections with multiple parasite strains complicate epidemiological and clinical assessments.

Purpose of the Study:

• To develop a robust statistical model for inferring the genetic structure of P. falciparum.
• To quantify the number of strains, their proportions, and unexplained mixture components from whole genome sequence (WGS) data.

Main Methods:

• A rigorous statistical model was developed and applied to simulation, laboratory, and field data.
• The model utilizes whole genome sequence (WGS) data to infer mixture composition.
• Performance was evaluated for accuracy and efficiency with varying data sizes.

Main Results:

• The model successfully infers P. falciparum mixture structure, including strain number and proportions.
• Accurate inference was achieved with minimal data, as few as 10 reads or 50 single nucleotide polymorphisms (SNPs).
• The model demonstrates efficiency with large datasets.

Conclusions:

• The developed statistical model provides a powerful tool for analyzing P. falciparum WGS data.
• It offers insights into within-host parasite dynamics, valuable for clinical and epidemiological studies.
• Open-source code and data facilitate broader application and research advancement.