Table 2

Summary of PROMPT's generic comparison methods and the corresponding examples presented. The symbol x in the data column means corresponding data values for the same protein, whereas a comma simply states that two sets of values are utilized.

Example

Type of data used

PROMPT method^a:

Applied statistical methods ^d

Fold comparison of GroEL substrates with the whole proteome

{ Nominal }, { Nominal }

Categorical feature comparison ^b

• Chi-Square test

Fold enrichment of GroEL substrates

{ Nominal }, subset of { Nominal }

Categorical feature enrichment ^c

• Sampling from hypergeometric distribution with correction

Abundance distribution of essential vs. all proteins

{ Numeric }, { Numeric }

Numeric distribution comparison

• Mann-Whitney (MW) and Kolmogorov-Smirnov (KS) of the whole distribution

• MW and Chi-Square test of each bin separately

Protein abundance vs. mRNA expression

{ Numeric × Numeric }

Numeric feature correlation

• Pearson correlation coefficient and

• Pearson correlation test

^a Extensive description of each method can be found in the context sensitive help integrated in the PROMPT GUI, or in the manual supplied with PROMPT.

^b Both groups with categorical data can be independent from each other.

^c One group must be drawn from the other group.

^d As described in the Methods section

Schmidt and Frishman BMC Bioinformatics 2006 7:331   doi:10.1186/1471-2105-7-331