The ability to distinguish between Escherichia coli strains is critical for outbreak investigations. Binary typing, based on the presence or absence of genetic material, provides a high-throughput alternative to geland PCR-based typing techniques that generate complex banding patterns and lack uniform interpretation criteria. We developed, validated, and determined the discriminatory power of an E. coli binary typing method, probe hybridization array typing (PHAT). In PHAT, the absence or presence of genetic material is identified by using DNA hybridization to produce a reproducible and portable fingerprint for each genome. PHAT probes were generated from genome subtractive hybridization experiments. We PHAT typed the ECOR collection of strains from a variety of geographical locations, and 33 rectal E. coli strains selected from college-aged women with urinary tract infection. In the set of 33 human rectal strains, the discriminatory power of PHAT (98%) equaled that of multilocus sequence typing (MLST) and pulsed-field gel electrophoresis. However, for ECOR strains, which include nonhuman strains, the current set of PHAT probes was less discriminating than MLST, ribotyping, and enterobacterial repetitive intergenic consensus sequence PCR (80% versus 97, 92, and 97%, respectively). When we limited the analysis to ECOR strains of B2 and D lineage, which are associated with human infection, current PHAT probes were highly discriminatory (94%). PHAT can be applied in a high-throughput format (i.e., "library on a slide"), the discriminatory ability can be varied based on the probe set, and PHAT is readily adapted to other bacterial species with high variation in genetic content.The ability to distinguish between Escherichia coli strains is critical for outbreak investigations; thus, the availability of rapid, reliable, valid, and high-throughput typing methods is desirable. Traditional serogroup-and phage-based typing methods have been increasingly replaced by more-rapid DNA fragment-based typing methods, including (i) repetitive sequence methods based on PCR such as enterobacterial repetitive intergenic consensus (ERIC) sequencing and randomly amplified polymorphic DNA (RAPD) detection (11,16,27), (ii) restriction digest and gel-based methods such as ribotyping and pulsed field gel electrophoresis (PFGE) (24), (iii) sequence-based methods such as multilocus sequence typing (MLST) (14, 24), (iv) whole-genome sequencing, and (v) single-nucleotide polymorphism (SNP) typing (10).Most gel-and PCR-based techniques generate complex banding patterns that lack uniform interpretation criteria (17). Although PFGE can be highly reproducible when a standard protocol and equipment is used, problems remain (17). The interpretation of gel-based methods is most straightforward when additional information regarding the relationships between strains is available, such as when they are epidemiologically linked and when assays are conducted in a single laboratory (24).DNA-based typing methods have the advantage of portability and reprod...