Skip to main content
Fig. 2 | Genome Biology

Fig. 2

From: A large-scale genomically predicted protein mass database enables rapid and broad-spectrum identification of bacterial and archaeal isolates by mass spectrometry

Fig. 2

Accuracy of taxonomic identification based on PM scores as evaluated with simulated peak lists for five representative genomes (Escherichia coli, GCF_003697165.2; Acinetobacter calcoaceticus, GCF_000368965.1; Bacillus subtilis, GCF_000009045.1; Cutibacterium acnes, GCF_003030305.1; Methanothermobacter thermautotrophicus, GCF_000008645.1). Peak matching was performed at a mass error tolerance of 200 ppm. Simulated peak lists consist of theoretically expected mass peaks containing at least six ribosomal proteins and varying amounts of random background peaks (noise). In brief, for each genome, level of noise, and the total number of peaks, 500 peak lists were generated, and the percentage of correct identification based on the best-matching hit was calculated at different taxonomic levels (genome level: red, species level: blue, genus level: green). The data are plotted as the means of all five genomes with error bars as the standard deviations. Panels A and B show the results based on scoring schemes II and III, respectively

Back to article page

Genome Biology

ISSN: 1474-760X

Contact us