Advanced

Peak Detection in Data Independent Acquisition Analysis

Stymne, Viktor LU (2016) In LU-CS-EX 2016-20 EDA920 20151
Department of Computer Science
Abstract
The analysis of peptides using mass spectrometry produces signals, where
intact and fragmented peptides are observable as specific signal peaks.
This report investigates how signal processing and supervised learning,
along with specialized peak detection algorithms, can improve the analysis of
such peptide signals in a mass spectrometry technique called Data Independent
Acquisition. To make this improvement possible, 1120 different possible
approaches were investigated and evaluated, and finally combined into an ensemble
system.
In this ensemble system, the best fit approach for data currently being processed
is chosen based on linear regression. The final system can identify all
the correct peaks from a manually-annotated test... (More)
The analysis of peptides using mass spectrometry produces signals, where
intact and fragmented peptides are observable as specific signal peaks.
This report investigates how signal processing and supervised learning,
along with specialized peak detection algorithms, can improve the analysis of
such peptide signals in a mass spectrometry technique called Data Independent
Acquisition. To make this improvement possible, 1120 different possible
approaches were investigated and evaluated, and finally combined into an ensemble
system.
In this ensemble system, the best fit approach for data currently being processed
is chosen based on linear regression. The final system can identify all
the correct peaks from a manually-annotated test set. (Less)
Please use this url to cite or link to this publication:
author
Stymne, Viktor LU
supervisor
organization
course
EDA920 20151
year
type
H3 - Professional qualifications (4 Years - )
subject
keywords
proteomics, signal processing, machine learning, peak detection
publication/series
LU-CS-EX 2016-20
report number
LU-CS-EX 2016-20
ISSN
1650-2884
language
English
id
8884133
date added to LUP
2016-06-22 12:49:57
date last changed
2016-06-22 12:49:57
@misc{8884133,
  abstract     = {The analysis of peptides using mass spectrometry produces signals, where
intact and fragmented peptides are observable as specific signal peaks.
This report investigates how signal processing and supervised learning,
along with specialized peak detection algorithms, can improve the analysis of
such peptide signals in a mass spectrometry technique called Data Independent
Acquisition. To make this improvement possible, 1120 different possible
approaches were investigated and evaluated, and finally combined into an ensemble
system.
In this ensemble system, the best fit approach for data currently being processed
is chosen based on linear regression. The final system can identify all
the correct peaks from a manually-annotated test set.},
  author       = {Stymne, Viktor},
  issn         = {1650-2884},
  keyword      = {proteomics,signal processing,machine learning,peak detection},
  language     = {eng},
  note         = {Student Paper},
  series       = {LU-CS-EX 2016-20},
  title        = {Peak Detection in Data Independent Acquisition Analysis},
  year         = {2016},
}