Unbiased Antimicrobial Resistance Detection from Clinical Bacterial Isolates Using Proteomics

Area of Science:

• Clinical microbiology
• Proteomics
• Antimicrobial resistance (AMR) detection

Background:

• Antimicrobial resistance (AMR) is a growing threat to bacterial infection treatment.
• Current AMR detection methods are slow and separate from species identification.
• Protein analysis is superior to gene or transcript sequencing for predicting AMR phenotypes.

Purpose of the Study:

• To develop an unbiased proteomics workflow for rapid bacterial species and AMR protein detection.
• To create a novel data analysis concept and software (rawDIAtect) for bacterial proteomics.
• To provide a solution for clinical microbiology needs, enabling faster AMR prediction.

Main Methods:

• An unbiased proteomics workflow was designed for primary cultures.
• The workflow detects bacterial species and AMR-related proteins within 4 hours.
• A new data analysis concept and software (rawDIAtect) were developed for peptide identification and AMR prediction.

Main Results:

• The method achieved 98% sensitivity and 100% specificity.
• Validated using 7 bacterial species and 11 AMR determinants (13 protein isoforms).
• Demonstrated proof-of-concept for rapid, accurate AMR detection.

Conclusions:

• The developed proteomics workflow enables rapid and accurate detection of bacterial species and AMR.
• The rawDIAtect software facilitates AMR prediction from peptide identifications.
• This approach significantly improves upon current methods for clinical microbiology settings.