Fig. 5: Experimental validation and potency assays of c_AMPs. | Nature Communications

Fig. 5: Experimental validation and potency assays of c_AMPs.

From: DLFea4AMPGen de novo design of antimicrobial peptides by integrating features learned from deep learning models

Fig. 5

a Inhibition rates of six bacterial and one fungal strain at 128 µM, and the scavenging rate of ABTS+ radicals at 1 mg/mL. Larger circles indicate better activity. Statistical significance was assessed using two-sided Dunnett’s test with correction for multiple comparisons: ‘*’ means 0.01 <p < 0.05; ‘**’ means 0.001 <p ≤ 0.01; and ‘***’ means p ≤ 0.001. b The hemolytic and cytotoxic activities of 16 c_AMPs were evaluated at a concentration of 128 µM. Each group included three biologically independent replicates (n = 3), with data presented as mean±s.d. Triton X-100 served as a positive control, while PBS was used as a negative control for hemolytic activity, and DMEM was the control for cytotoxicity. The growth curves of representative drug-resistant bacteria, which were E. coli z44 (c) isolated from sick chickens in livestock and S. aureus 09057 (d) from clinical environments. Each group included three biologically independent replicates (n = 3), with data presented as mean ± s.d. The control group demonstrated normal bacterial growth. The concentration gradient of D1 and D2 was set from 128 µM to 1 µM in two-fold dilutions, with different colors representing each concentration. The concentration indicated in each panel is the MIC of the peptide against each bacterial strain. e SEM images of S. aureus 09057 treated with PBS and D1 (64 μM) at low magnification (scale bar, 2 μm). f TEM images of S. aureus 09057 treated with PBS and D1 (64 μM) at low magnification (scale bar, 200 nm). Images shown are representative of at least three independent experiments with similar results. Additional images and magnifications are provided in Supplementary Fig. 15, confirming the reproducibility of the observations.

Back to article page