Difference between revisions of "Review, evaluation, and discussion of the challenges of missing value imputation for mass spectrometry-based label-free global proteomics."

(Created page with " == General information == (13) Webb-Robertson, B.-J. M.; Wiberg, H. K.; Matzke, M. M.; Brown, J. N.; Wang, J.; McDermott, J. E.; Smith, R. D.; Rodland, K. D.; Metz, T. O.; Po...")
 
(Citation)
 
(7 intermediate revisions by the same user not shown)
Line 1: Line 1:
  
== General information ==
+
=== Citation ===
(13) Webb-Robertson, B.-J. M.; Wiberg, H. K.; Matzke, M. M.;
+
Webb-Robertson, B.-J. M.; Wiberg, H. K.; Matzke, M. M.;
 
Brown, J. N.; Wang, J.; McDermott, J. E.; Smith, R. D.; Rodland, K. D.;
 
Brown, J. N.; Wang, J.; McDermott, J. E.; Smith, R. D.; Rodland, K. D.;
Metz, T. O.; Pounds, J. G.; Waters, K. M.; et al. Review, evaluation,
+
Metz, T. O.; Pounds, J. G.; Waters, K. M.; et al. [https://doi.org/10.1371/journal.pone.0074335 Review, evaluation, and discussion of the challenges of missing value imputation for mass spectrometry-based label-free global proteomics.] J. Proteome Res. 2015,
and discussion of the challenges of missing value imputation for mass
 
spectrometry-based label-free global proteomics. J. Proteome Res. 2015,
 
 
14 (5), 1993−2001.
 
14 (5), 1993−2001.
[https://doi.org/10.1371/journal.pone.0074335: DOI]
 
  
 +
[https://doi.org/10.1371/journal.pone.0074335 Permanent link to the article]
  
 
=== Summary ===
 
=== Summary ===
Line 14: Line 12:
  
 
=== Study outcomes ===
 
=== Study outcomes ===
List the paper results concerning method comparison and benchmarking:
 
 
==== Outcome O1 ====
 
==== Outcome O1 ====
 
Most imputation methods perform well, no single algorithm or imputation strategy (single, local, global) outperforms, sometimes even no imputation is superior in subsequent classification analysis.
 
Most imputation methods perform well, no single algorithm or imputation strategy (single, local, global) outperforms, sometimes even no imputation is superior in subsequent classification analysis.
  
 
==== Outcome O2 ====
 
==== Outcome O2 ====
Local similarity-based approaches are in general the most accuarate and robust methods. Such as least-squares adaptive (LSA) or regularized expectation maximization (REM).
+
Local similarity-based approaches are in general the most accuarate and robust methods. Such as least-squares adaptive (LSA) or regularized expectation maximization (REM) (Figure 4)
  
 
==== Outcome O3 ====
 
==== Outcome O3 ====
With left-censored data the number of missing values highly depends on peptide intensity (Figure 1)
+
The 'best' imputation method highly depends on the data and the goal of the downstream analysis and therewith advantageous methods are hard to define (Figure 3)
  
 
==== Further outcomes ====
 
==== Further outcomes ====
The 'best' imputation method highly depends on the data and the goal of the downstream analysis and therewith advantageous methods are hard to define.
+
With left-censored data the number of missing values highly depends on peptide intensity (Figure 1)
 
 
  
 
=== Study design and evidence level ===
 
=== Study design and evidence level ===
Line 32: Line 28:
 
3 single-value approaches (LOD1,LOD2,RTI), 5 local similarity approaches (KNN, LLS, LSA, REM, MBI) and 2 global-structure approaches (PPCA, BPCA) were evaluated which allows comparison and discussion of different imputation strategies.
 
3 single-value approaches (LOD1,LOD2,RTI), 5 local similarity approaches (KNN, LLS, LSA, REM, MBI) and 2 global-structure approaches (PPCA, BPCA) were evaluated which allows comparison and discussion of different imputation strategies.
 
They were applied to 3 real datasets of different type and species, which represent a broad biological application.
 
They were applied to 3 real datasets of different type and species, which represent a broad biological application.
 
  
 
=== Further comments and aspects ===
 
=== Further comments and aspects ===
  
 
=== References ===
 
=== References ===
The list of cited or related literature is placed here.
 

Latest revision as of 11:49, 25 February 2020

Citation

Webb-Robertson, B.-J. M.; Wiberg, H. K.; Matzke, M. M.; Brown, J. N.; Wang, J.; McDermott, J. E.; Smith, R. D.; Rodland, K. D.; Metz, T. O.; Pounds, J. G.; Waters, K. M.; et al. Review, evaluation, and discussion of the challenges of missing value imputation for mass spectrometry-based label-free global proteomics. J. Proteome Res. 2015, 14 (5), 1993−2001.

Permanent link to the article

Summary

Evaluation of performance and caveats of 9 imputation algorithms applied on a LC-MS data set.

Study outcomes

Outcome O1

Most imputation methods perform well, no single algorithm or imputation strategy (single, local, global) outperforms, sometimes even no imputation is superior in subsequent classification analysis.

Outcome O2

Local similarity-based approaches are in general the most accuarate and robust methods. Such as least-squares adaptive (LSA) or regularized expectation maximization (REM) (Figure 4)

Outcome O3

The 'best' imputation method highly depends on the data and the goal of the downstream analysis and therewith advantageous methods are hard to define (Figure 3)

Further outcomes

With left-censored data the number of missing values highly depends on peptide intensity (Figure 1)

Study design and evidence level

General aspects

3 single-value approaches (LOD1,LOD2,RTI), 5 local similarity approaches (KNN, LLS, LSA, REM, MBI) and 2 global-structure approaches (PPCA, BPCA) were evaluated which allows comparison and discussion of different imputation strategies. They were applied to 3 real datasets of different type and species, which represent a broad biological application.

Further comments and aspects

References