Table 2 A comparison of the strengths and weaknesses of in silico gene expression mining tools

Strengths:

Size of EST databases in Unigene

Conservative test (Fisher's exact test) used to determine significance

Absolute and relative counts given

Weakness:

Libraries with low EST count excluded by analysis

Limited number of "normal tissue" libraries

DGED

Strengths:

Statistically parameters can be varied

Results linked to tissue microarray data

Ability to select origin/type of tissue (e.g. micro dissected etc).

Genes with low abundance included

Weakness:

Comparison based on odds ratio

Sagemap

Strengths:

Wide variety in the source of SAGE data available.

Accounts for differences in sample size between groups

Weakness:

Exclusion of tags with low counts

XProfiler

Strengths:

Ability to compare groups and pools of libraries.

Outputs genes as unique/non-unique and known/unknown.

Ability to select origin/type of tissue (e.g. micro dissected etc).

Weakness:

Exclusion of tags with low counts

Common Strengths:

Freely available via internet

Unbiased view of transcriptome

Common Weaknesses:

Reliability of initial sequencing experiments.

Limited background knowledge of original tissues

Significant false positive rate/false negative rate unknown