Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Chromatographic retention time prediction and its applications in mass spectrometry-based proteomics
Stockholm University, Faculty of Science, Department of Biochemistry and Biophysics. (Lukas Käll)ORCID iD: 0000-0002-5034-5379
2013 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Mass spectrometry-based methods are among the most commonly used techniques to characterize proteins in biological samples. With rapid technological developments allowing increasing throughput, thousands of proteins can now be monitored in a matter of hours. However, these advances brought a whole new set of analytical challenges. At the moment, it is no longer possible to rely on human experts to process the data. Instead, accurate computational tools are required.

In line with these observations, my research work has involved development of computational methods to facilitate the analysis of mass spectrometry-based experiments. In particular, the projects included in this thesis revolve around the chromatography step of such experiments, where peptides are separated according to their hydrophobicity.

The first part of the thesis describes an algorithm to predict retention time from peptide sequences. The method provides more accurate predictions compared to previous approaches, while being easily transferable to other chromatography setups. In addition, it gives equally good predictions for peptides carrying arbitrary posttranslational modifications as for unmodified peptides.

The second part of the thesis includes two applications of retention time predictions in the context of mass spectrometry-based proteomics experiments. First, we show how theoretical calculations of masses and retention times can be used to infer proteins in shotgun proteomics experiments. Secondly, we illustrate the use of retention time predictions to calculate optimized gradient functions for reversed-phase liquid chromatography.

Place, publisher, year, edition, pages
Department of Biochemistry and Biophysics. Stockholm University , 2013. , 58 p.
Keyword [en]
mass spectrometry; reversed-phase liquid chromatography; retention time prediction; bioinfomatics; computational mass spectrometry
National Category
Bioinformatics and Systems Biology
Research subject
Biochemistry towards Bioinformatics
Identifiers
URN: urn:nbn:se:su:diva-94800ISBN: 978-91-7447-775-7 (print)OAI: oai:DiVA.org:su-94800DiVA: diva2:656055
Public defence
2013-11-29, Magnelisalen, Kemiska övningslaboratoriet, Svante Arrhenius väg 16 B, Stockholm, 10:00 (English)
Opponent
Supervisors
Note

A the time of the doctoral defence the paper nr. 4 had a status Epubl. ahead of  print.

Available from: 2013-11-07 Created: 2013-10-14 Last updated: 2013-10-23Bibliographically approved
List of papers
1. Training, Selection, and Robust Calibration of Retention Time Models for Targeted Proteomics
Open this publication in new window or tab >>Training, Selection, and Robust Calibration of Retention Time Models for Targeted Proteomics
2010 (English)In: Journal of Proteome Research, ISSN 1535-3893, E-ISSN 1535-3907, Vol. 9, no 10, 5209-5216 p.Article in journal (Refereed) Published
Abstract [en]

Accurate predictions of peptide retention times (RT) in liquid chromatography have many applications in mass spectrometry-based proteomics. Most notably such predictions are used to weed out incorrect peptide-spectrum matches, and to design targeted proteomics experiments. In this study, we describe a RT predictor, ELUDE, which can be employed in both applications. ELUDE's predictions are based on 60 features derived from the peptide's amino acid composition and optimally combined using kernel regression. When sufficient data is available, ELUDE derives a retention time index for the condition at hand making it fully portable to new chromatographic conditions. In cases when little training data is available, as often is the case in targeted proteomics experiments, ELUDE selects and calibrates a model from a library of pretrained predictors. Both model selection and calibration are carried out via robust statistical methods and thus ELUDE can handle situations where the calibration data contains erroneous data points. We benchmarked our method against two state-of-the-art predictors and showed that ELUDE outperforms these methods and tracked up to 34% more peptides in a theoretical SRM method creation experiment. ELUDE is freely available under Apache License from http://per-colator.com.

Keyword
retention time prediction, support vector regression, targeted proteomics, peptide identification, bioinformatics
National Category
Natural Sciences
Identifiers
urn:nbn:se:su:diva-49432 (URN)10.1021/pr1005058 (DOI)000282257800031 ()
Note
authorCount :3Available from: 2010-12-17 Created: 2010-12-14 Last updated: 2017-12-11Bibliographically approved
2. Chromatographic retention time prediction for posttranslationally modified peptides
Open this publication in new window or tab >>Chromatographic retention time prediction for posttranslationally modified peptides
Show others...
2012 (English)In: Proteomics, ISSN 1615-9853, E-ISSN 1615-9861, Vol. 12, no 8, 1151-1159 p.Article in journal (Refereed) Published
Abstract [en]

Retention time prediction of peptides in liquid chromatography has proven to be a valuable tool for mass spectrometry-based proteomics, especially in designing more efficient procedures for state-of-the-art targeted workflows. Additionally, accurate retention time predictions can also be used to increase confidence in identifications in shotgun experiments. Despite these obvious benefits, the use of such methods has so far not been extended to (posttranslationally) modified peptides due to the absence of efficient predictors for such peptides. We here therefore describe a new retention time predictor for modified peptides, built on the foundations of our existing Elude algorithm. We evaluated our software by applying it on five types of commonly encountered modifications. Our results show that Elude now yields equally good prediction performances for modified and unmodified peptides, with correlation coefficients between predicted and observed retention times ranging from 0.93 to 0.98 for all the investigated datasets. Furthermore, we show that our predictor handles peptides carrying multiple modifications as well. This latest version of Elude is fully portable to new chromatographic conditions and can readily be applied to other types of posttranslational modifications. Elude is available under the permissive Apache2 open source License at or can be run via a web-interface at .

Keyword
Bioinformatics, Machine learning, Posttranslational modification, Retention time prediction, Reversed-phase liquid chromatography
National Category
Medical Biotechnology
Identifiers
urn:nbn:se:su:diva-80742 (URN)10.1002/pmic.201100386 (DOI)000303918200009 ()
Note

AuthorCount:7;

Available from: 2012-10-01 Created: 2012-09-27 Last updated: 2017-12-07Bibliographically approved
3. Optimized Nonlinear Gradients for Reversed-Phase Liquid Chromatography in Shotgun Proteomics
Open this publication in new window or tab >>Optimized Nonlinear Gradients for Reversed-Phase Liquid Chromatography in Shotgun Proteomics
Show others...
2013 (English)In: Analytical Chemistry, ISSN 0003-2700, E-ISSN 1520-6882, Vol. 85, no 16, 7777-7785 p.Article in journal (Refereed) Published
Abstract [en]

Reversed-phase liquid chromatography has become the preferred method for separating peptides in most of the mass spectrometry-based proteomics workflows of today. In the way the technique is typically applied, the peptides are released from the chromatography column by the gradual addition of an organic buffer according to a linear function. However, when applied to complex peptide mixtures, this approach leads to unequal spreads of the peptides over the chromatography time. To address this, we investigated the use of nonlinear gradients, customized for each setup at hand. We developed an algorithm to generate optimized gradient functions for shotgun proteomics experiments and evaluated it for two data sets consisting each of four replicate runs of a human complex sample. Our results show that the optimized gradients produce a more even spread of the peptides over the chromatography run, while leading to increased numbers of confident peptide identifications. In addition, the list of peptides identified using nonlinear gradients differed considerably from those found with the linear ones, suggesting that such gradients can be a valuable tool for increasing the proteome coverage of mass spectrometry-based experiments.

National Category
Analytical Chemistry
Identifiers
urn:nbn:se:su:diva-94188 (URN)10.1021/ac401145q (DOI)000323471800024 ()
Note

AuthorCount:5;

Funding Agencies:

Science for Life Laboratory;  European Commission;   Austrian Science Fund via the Special Research Program Chromosome Dynamics  SFBF3402; Translational-Research-Program  TRP308 

Available from: 2013-10-01 Created: 2013-09-30 Last updated: 2017-12-06Bibliographically approved
4. Mass fingerprinting of complex mixtures: protein inference from high-resolution peptide masses and predicted retention times
Open this publication in new window or tab >>Mass fingerprinting of complex mixtures: protein inference from high-resolution peptide masses and predicted retention times
Show others...
2013 (English)In: Journal of Proteome Research, ISSN 1535-3893, E-ISSN 1535-3907, Vol. 12, no 12, 5730-5741 p.Article in journal (Refereed) Published
Abstract [en]

In typical shotgun experiments, the mass spectrometer records the masses of a large set of ionized analytes but fragments only a fraction of them. In the subsequent analyses, normally only the fragmented ions are used to compile a set of peptide identifications, while the unfragmented ones are disregarded. In this work, we show how the unfragmented ions, here denoted MS1-features, can be used to increase the confidence of the proteins identified in shotgun experiments. Specifically, we propose the usage of in silico mass tags, where the observed MS1-features are matched against de novo predicted masses and retention times for all peptides derived from a sequence database. We present a statistical model to assign protein-level probabilities based on the MS1-features and combine this data with the fragmentation spectra. Our approach was evaluated for two triplicate data sets from yeast and human, respectively, leading to up to 7% more protein identifications at a fixed protein-level false discovery rate of 1%. The additional protein identifications were validated both in the context of the mass spectrometry data and by examining their estimated transcript levels generated using RNA-Seq. The proposed method is reproducible, straightforward to apply, and can even be used to reanalyze and increase the yield of existing data sets.

Place, publisher, year, edition, pages
American Chemical Society (ACS), 2013
Keyword
mass spectrometry, shotgun proteomics, mass fingerprinting
National Category
Biochemistry and Molecular Biology
Identifiers
urn:nbn:se:su:diva-94614 (URN)10.1021/pr400705q (DOI)000328231300033 ()
Note

AuthorCount: 6

Funding Agencies:

Swedish Research Council;   U.S. National Science Foundation (MRI)  0923536; American Recovery and Reinvestment Act (ARRA) funds  R01 HG005805;  National Institute of General Medical Sciences/Center for Systems Biology  2P50 GM076547;  Luxembourg Centre for Systems Biomedicine;   University of Luxembourg  

Available from: 2013-10-07 Created: 2013-10-07 Last updated: 2017-12-06Bibliographically approved

Open Access in DiVA

No full text

Search in DiVA

By author/editor
Moruz, Luminita
By organisation
Department of Biochemistry and Biophysics
Bioinformatics and Systems Biology

Search outside of DiVA

GoogleGoogle Scholar

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 262 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf