CRISPR Array Reconstruction Shields Light on Diversity and Dynamics of Escherichia coli Populations

Genotyping of Escherichia coli isolates is indispensable for clinical diagnostics and epidemiological studies. MLVA (Multiple locus variable number tandem repeat), SNP (single nucleotide polymorphism) and antibiotic resistance spectrum analyzes are not sufficient to meet the needs for fast and discriminative approach. The diversity of CRISPR arrays may constitute a basis for E. coli genotyping. Here, we develop an approach based on machine learning. It uses short fragments of CRISPR arrays amplified with primers complementary to CRISPR repeats to reconstruct whole length arrays. The procedure does not need individual clone isolation and is applicable to complex microbiotas. In addition, E. coli clonal diversity and dynamics in intestinal samples revealed with the new approach will be discussed.