Difference between revisions of "Missing value estimation methods for DNA microarrays"

(Created page with " === Citation === Troyanskaya, O., Cantor, M., Sherlock, G., Brown, P., Hastie, T., Tibshirani, R., Botstein, D., and Altman, R. B. (2001). Missing value estimation methods fo...")
 
 
(2 intermediate revisions by the same user not shown)
Line 1: Line 1:
 +
__NUMBEREDHEADINGS__
  
 
=== Citation ===
 
=== Citation ===
Line 4: Line 5:
 
Botstein, D., and Altman, R. B. (2001). Missing value estimation methods for dna
 
Botstein, D., and Altman, R. B. (2001). Missing value estimation methods for dna
 
microarrays. Bioinformatics, 17(6):520–525.
 
microarrays. Bioinformatics, 17(6):520–525.
[https://doi.org/10.1093/bioinformatics/17.6.520: 10.1093/bioinformatics/17.6.520]
+
[https://doi.org/10.1093/bioinformatics/17.6.520 Permanent link to paper]
  
 
=== Summary ===
 
=== Summary ===
Line 10: Line 11:
  
 
=== Study outcomes ===
 
=== Study outcomes ===
List the paper results concerning method comparison and benchmarking:
 
 
==== Outcome O1 ====
 
==== Outcome O1 ====
 
Rank of performance: KNN, SVD, row average, zero filling.
 
Rank of performance: KNN, SVD, row average, zero filling.

Latest revision as of 15:22, 25 February 2020


1 Citation

Troyanskaya, O., Cantor, M., Sherlock, G., Brown, P., Hastie, T., Tibshirani, R., Botstein, D., and Altman, R. B. (2001). Missing value estimation methods for dna microarrays. Bioinformatics, 17(6):520–525. Permanent link to paper

2 Summary

SVD, KNN and row average imputation are evaluated with different parameter settings on real data sets with regard to robustness, sensitivity and accuracy.

3 Study outcomes

3.1 Outcome O1

Rank of performance: KNN, SVD, row average, zero filling.

3.2 Outcome O2

"KNN is relatively insensitive to .. K within the range of k=10-20" (Figure 1)

3.3 Outcome O3

SVD "is sensitive to the type of data" and "is ideally suited .. in terms of .. constituent patterns"

4 Study design and evidence level

Just 4 imputation algorithms (SVD,KNN) are evaluated from which 2 are singular value substitutions (average,zero).

Analysis is performed over a broad range of hyperparameters (KNN: k=[1,1000], SVD: Eigengenes=[5,30]).

The imputation methods are only analyzed on data with less than 20%.

5 References