Method for identifying microorganisms by mass spectrometry
Inventors
Strubel, Grégory • Arsac, Maud • Desseree, Denis • Cotte-Pattat, Pierre-Jean
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12191129-B2
Publication Date
2025-01-07
Expiration Date
Abstract
A method of identifying a microorganism by mass spectrometry, including acquiring at least one mass spectrum of said microorganism; for each acquired mass spectrum: detecting peaks of the spectrum in a predetermined mass range; generating a list of peaks identifying at most one peak in each interval of a predetermined subdivision of the range of mass-to-charge ratios, the width of the intervals of the subdivision logarithmically increasing along with the mass-to-charge ratio, and analyzing the list(s) of peaks obtained according to a knowledge base of previously-identified microorganisms and/or types of microorganisms.
Core Innovation
The invention relates to providing an identification decision for a microorganism using a processor-based calculation unit and a mass spectrum obtained by mass spectrometry from a mass spectrometer. At least one mass spectrum is provided for a sample of the microorganism over a predetermined range of mass-to-charge ratios, and peaks are detected within the predetermined range.
A list of peaks is generated using intervals from a predetermined subdivision of the range, where the interval width increases along with the mass-to-charge ratio according to relations using mmin and mmax and integer-indexed intervals. For each acquired mass spectrum, the processor identifies at most one peak in each interval to obtain a vector, and the vector(s) form a matrix representative of the microorganism.
The representation is based on peak retention per interval, including retaining the maximum peak in the interval, and the matrix can be binarized to indicate presence or absence of peaks within intervals. An identification decision is provided by comparing the representative matrix with a knowledge base of matrices representative of previously-identified microorganisms and/or types of microorganisms stored in memory.
The approach is described as reducing peak-position uncertainty and reducing data size by using logarithmically increasing interval widths with at most one peak per interval. Comparative tests are described as showing lower identification error and reduced occupied memory versus constant quantization.
Claims Coverage
The independent claim describes three core inventive portions: detect peaks over a predetermined m/z range, generate interval-dependent peak lists using intervals whose width increases with m/z so that at most one peak per interval yields a vector and matrix representation, and provide an identification decision by comparing the representative matrix to a stored knowledge base of matrices for previously-identified microorganisms/types.
Processor-based microorganism identification decision using representative matrix comparison
Providing an identification decision for the microorganism by comparing a representative matrix formed from vectors derived from mass spectra with a knowledge base of matrices representative of previously-identified microorganisms and/or types of microorganisms, wherein the knowledge base is stored in a memory.
Peak detection over a predetermined mass-to-charge range and interval-dependent peak list
Detecting peaks of the mass spectrum in the predetermined range of mass-to-charge ratios, and generating a list of peaks in intervals of a predetermined subdivision of the predetermined range of mass-to-charge ratios, wherein the width of the intervals increases along with the mass-to-charge ratio according to relations using mmin and mmax.
At most one peak per interval to form a vector and matrix representative of the microorganism
Identifying at most one peak in each interval in the list of peaks so as to obtain a vector for each of the at least one mass spectrum, wherein the vector(s) form a matrix representative of the microorganism.
Across the independent claim, identification is achieved by converting mass-spectrometry peak information into an interval-dependent vector/matrix representation with at most one peak per interval and increasing interval widths across m/z, and then performing a matrix comparison against a stored knowledge base of previously identified microorganisms/types.
Stated Advantages
Reduced peak-position uncertainty.
Reduced data size.
Lower identification error versus constant quantization.
Reduced occupied memory versus constant quantization.
Documented Applications
Providing an identification decision for a microorganism based on a mass spectrum obtained by mass spectrometry, including MALDI-TOF.
Interested in licensing this patent?