|Version 42 (modified by v.mitra@…, 4 years ago) (diff)|
Welcome to project Pre-alignment
The pre-alignment based warping of LC-MS peak-lists allows direct comparison of complex LC-MS datasets that have been sourced in an inter-laboratory study design or are highly variable. The tool has been developed primarily for time alignment purposes; hence vital pre-processing steps should be carried out before alignment of peak-lists. The tool aligns the retention time axes of multiple input files and corrects large non-linear retention time shifts in complex LC-MS datasets.
Major Application:Reliable biomarker discovery based on profiling of complex body fluids requires routine analysis of large number of samples and demands LC-MS platforms with high sample throughput. Inter-laboratory sample comparisons can increase this throughput and can also confirm the reliability of data produced at different analysis centers. However due to large retention time shifts that often occur in these inter-laboratory datasets; state of the art time alignment algorithms are unable to correct these retention differences. The current work proposes a novel two step time alignment strategy for LC-MS datasets obtained in an inter-laboratory experimental design. The method composes a robust pre-alignment function that is generated by identifying, with suitable accuracy, true peak correspondences between highly variable chromatograms, followed by a more accurate time alignment using the Warp2D algorithm [(F. Suits et al.; 2008,http://pubs.acs.org/doi/abs/10.1021/ac702267h)].
To read how to install please click - InstallGuide
pre_align:The pre-alignment based warping of LC-MS peak-lists allows direct comparison of complex LC-MS datasets that have been sourced in an inter-laboratory study design or are highly variable. Usage: pre_align <options> Options : -in <refdir> <sampdir> Path of Input files (valid formats: 'mzXML') -in <MCRpath> Path of the deployed MCR installation -in_param <param> Total number of peaks to be used between peaklist to produce a trend, delta mass tolerance for matching peaks in daltons eg: 0.1,0.2[[BR]],Size of sliding Rank window parameter. -out <dir> Path for warped files
java -jar pre_align.java (click here)!
This project provides the following mailing lists.
- pre-alignment-users: a list intended for general discussion on the project.
- pre-alignment-commits: a list that receives source code commit messages.
- pre-alignment-devel: a list intended for discussion among developers (subscription is restricted to registered developers).
If available, anonymous readonly subversion access works as follows:
svn co https://trac.nbic.nl/svn/pre-alignment pre-alignment
Write access is only available to registered developers.
Acronyms and Definition
- LC-MS Peaklists: A peak list contains information about peaks that have been observed over a small set of raw data points and are above a threshold local maximum of the ion count (intensity) in the LC-MS spectra.
- Bi-Variate Kernel Density estimation: A non-parametric approach of estimating density values using probability density functions, having a 2 dimensional Gaussian kernel or any other kernel. The data-points that are used in this approach are defined by a pair of variables.
- LOWESS: Local regression using weighted linear least squares and a 1st degree polynomial model.
- PCHIP: Piecewise Cubic Hermite Interpolating Polynomial; a cubic curve fitting spline function which preserves monotonicity of the data http://math.lanl.gov/~mac/papers/numerics/H83.pdf
- Warp2D: A time alignment method that takes advantage of both dimensions of LC-MS data to resolve ambiguities in peak matching. http://pubs.acs.org/doi/full/10.1021/ac702267h
- TracGuide -- Built-in Documentation
- TitleIndex -- A complete list of local wiki pages.
- Trac FAQ -- Frequently Asked Questions
Other NBIC software projects
All active NBIC software projects can be accessed from the project index.