Fig. 5: Nanomotion AST for strongly bactericidal CZA based on 1-SP classification. | Nature Communications

Fig. 5: Nanomotion AST for strongly bactericidal CZA based on 1-SP classification.

From: Accurate and rapid antibiotic susceptibility testing using a machine learning-assisted nanomotion technology platform

Fig. 5

a Nanomotion recordings for three E.coli isolates: the susceptible (S) ATCC-25922 (left), the resistant to ceftazidime, CAZ, but susceptible to ceftazidime-avibactam, CZA, (R/S) IHMA-2155385 (middle), and the resistant (R) BAA-2452 (right). Strains were exposed to 32 μg/ml of CAZ either alone (blue) or in combination with 4 μg/ml of the β-lactamase inhibitor avibactam, AVI (red). The median variance of three recordings is shown. b SP “slope” for all three strains exposed a constant concentration of 4 μg/ml avibactam (AVI) as well as increasing concentrations of ceftazidime (CAZ). Each data point depicts the median of three experiments, with the shaded area marking the standard error. c MIC distribution of E. coli strains used to develop a CZA classification model. d A single-SP model perfectly discriminates all susceptible (S) and resistant (R) E. coli strains; nanomotion AST was performed on cells treated with a combination of ceftazidime (32 μg/ml) and avibactam (4 μg/ml). A two-tailed Mann–Whitney U test was used for statistical analysis, p ≤ 0.0001. Data are available in the source data file. Single SP values and scores are available in Supplementary Data 2.

Back to article page