Systems and methods for the automated diagnosis of mycobacterial infection

A method for diagnosing a mycobacterial infection in a subject, the method comprising:

• performing LC-MS/MS and bottom-up proteomic analysis on an enzymatically digested HMW depleted CFP sample from the subject and identify a plurality of sample peptides;
• analyzing the plurality of sample peptides with a classifier algorithm, wherein the classifier algorithm comprises:
• a first section, wherein the first section is configured to analyze and cross reference each of the plurality of sample peptides to a database and identify all species of organisms for which each of the plurality of sample peptides is associated;
• an identification section, wherein the identification section is configured analyze the output for the first section and eliminate all peptides not associated with a mycobacteria species and to output a refined plurality of sample peptides;
• a scoring section, wherein the scoring section analyzes the output of the identification section and eliminates all peptides that are not specific to a single mycobacteria species, either alone or combined with a second or third sample peptide and outputting a plurality of species specific sample peptides;
• and
• wherein the classifier algorithm is further configured to compile the output of the scoring section and calculate the PWsp score by dividing the number of species-specific sample peptides (output of the scoring section) by the number of peptides in the refined plurality of sample peptides (output of the identification section).

A system for diagnosing a mycobacterial infection in a subject comprising:

• a module, configured to perform acetonitrile fractionation of a culture filtrate protein (CFP) sample obtained from the subject and retaining the supernatant fraction, wherein the retained supernatant fraction is selectively depleted in high molecular weight (HMW) proteins and to perform an enzymatic digest of the supernatant fraction to produce a digested peptide sample;
• a module configured to perform liquid chromatography tandem mass spectrometry (LC-MS/MS) and bottom-up proteomic analysis of the digested peptide sample to identify a plurality of sample peptides;
• one or more processors configured to:
• analyze the plurality of sample peptides with a classifier algorithm, wherein the classifier algorithm is configured to cross reference each of the plurality of sample peptides with a mycobacterium peptide database to determine which of the plurality of sample peptides, individually or in combination, are specific to a mycobacterium species or subspecies in the database;
• and
• generate a PW species identification score (PWsp) by dividing the number of single and multi-peptide combinations that identify a specific mycobacterium species by the total number of peptide combinations specific for any mycobacteria, wherein a higher (PWsp) for a given species of mycobacterium indicates a higher likelihood the subject has an infection of that species.