CoPTR-Contig is accurate on simulated and real data. (A) Comparison of CoPTR-Ref and CoPTR-Contig on simulated data using the E. coli density map. Performance was evaluated by computing the correlation (y-axis) between simulated and estimate log2(PTR)s across read counts (x-axis), randomly chosen replication origins, and PTRs. CoPTR-Contig shows high accuracy above 5000 reads. (B) Comparison of CoPTR-Contig to GRiD, DEMIC, and iRep across five genomic (monoculture) and metagenomic data sets (x-axis). For each data set, reads were mapped to a single reference genome for each species (see Methods). Performance was evaluated by comparing log2(PTR) estimates from CoPTR-Ref to the log2(PTR) estimate from each method across 10 high-quality metagenome assembled genomes (MAGs; points on the figure). Significance was computed using a two-tailed t-test: (*) P < 0.05, (**) P < 10−2, (***) P < 10−3, (****) P < 10−4. (C) Number of PTR estimates from species passing the filtering criteria for each model. The mean and SD are reported for the E. coli metagenomic gut and genomic data sets across MAGs from B. Error bars depict 1 SD. Each model was also applied to 10 samples from the IBD data set using 1009 high-quality MAGs from the IGGdb. The total number of PTRs passing filtering criteria for each model is reported.
