SIEVE User Guide

A Understanding the ChromAlign Process
Outlining the ChromAlign Process
Outlining the ChromAlign Process
Start with Original Chromatograms
A and B
Step 1
Align chromatograms roughly using Fast
Fourier Transform (FFT) algorithm
Step 2
Generate correlation matrix, (C (A, B))
Step 3
Create dynamic programming
optimal path in C(A, B - ) and provide a
final alignment
After selecting samples and determining the control data set, send the
control and sample chromatograms to be aligned with respect to the first
control chromatogram.
Step 1 of the ChromAlign process produces a crude alignment of the two
chromatographic surfaces using a fast fourier transform algorithm to
maximize the overlap of the chromatographic surfaces:
FFT (A) (first control sample)
FFT (B) (all other chromatograms)
FFT -1 (FFT (B)xFFT (A)*)
Crudely aligned, B -
In step 2, SIEVE uses the full MS scan information to fully align all
components (major as well as minor components of the samples) of the
chromatographic surfaces. SIEVE uses dynamic programming to maximize
the sum of all correlation co-efficients of peaks between time 0 and the end
of the chromatogram.
SIEVE completely aligns the chromatographic surfaces.
28 SIEVE User Guide Thermo Fisher