MalKinID: A classification model for identifying malaria parasite genealogical relationships using...

Area of Science:

• Genomics
• Parasitology
• Computational Biology

Background:

• Pathogen genomics is crucial for tracking infectious disease transmission.
• Identity-by-descent (IBD) analysis assesses genetic relatedness in malaria parasites.
• IBD can reconstruct transmission history and identify parasites for quantitative trait locus (QTL) experiments.

Purpose of the Study:

• To introduce MalKinID (Malaria Kinship Identifier), a novel classification model for identifying genealogical relationships among malaria parasites.
• To evaluate MalKinID's performance in distinguishing parent-child and sibling relationships based on IBD.
• To assess MalKinID's utility in reconstructing transmission history under various scenarios, including outcrossing and inbreeding.

Main Methods:

• Developed MalKinID, a model utilizing genome-wide identity-by-descent proportions and segment distributions.
• Calibrated MalKinID using genomic data from three laboratory-based genetic crosses.
• Tested MalKinID on simulated outcrossed and inbred malaria parasite importations.

Main Results:

• MalKinID achieved >80% sensitivity in identifying laboratory-generated F1 progeny.
• The model accurately identified F1 and backcross progeny in a specific cross.
• MalKinID demonstrated high precision and sensitivity in reconstructing outcrossed genealogies (F1-scores > 0.84).
• Performance declined with inbred relationships, necessitating multisample comparisons for accurate inference.

Conclusions:

• MalKinID provides a robust framework for using identity-by-descent to track malaria parasite transmission.
• The model is effective for separating parasite progeny for QTL experiments.
• Accurate genealogical inference is enhanced by outcrossing or employing multisample comparisons with predefined pedigrees.