Expander version 2 - paper version: ---------------------------------------- Changes and improvements from 1.2.1: 1) An advanced option has been added to the expression data load dialog box. This option enables the user to set several parameters regarding the input files and to view the first few rows as they will be read according to the input. This utility significantly increases the robustness and flexibility of the data load. 2) A flooring option has been added, and can be used in the data load stage, via the "Advanced" dialog box. 3) An option to save the data into a text file at any stage has been added, and can be operated by selecting "File">> "Save preprocessed data". 4) Fingerprint and annotation files have been updated and are now available for mouse, human, yeast, rat, fly and worm. 5) An option to configure the region (bounds and coding/non coding regions) on which promoter analysis is performed has been added to the Promoter Analysis settings dislog box. 6) The functional analysis utility has been improved and is now performed by the TANGO software. This software uses pre-processed tables of genes and their GO annotation, and perform hyper-geometric enrichment tests for sets of genes with common annotation and sets of genes given as input. Importantly, TANGO correct for multiple testing at both the multiple GO classes and multiple tested sets levels. It does so by bootstrapping and estimating the empirical p-value distribution for the evaluated sets. The resulting visualizations display both the corrected and the uncorrected p-values. 7) Histogram column heights in the functional analysis and the promoter analysis displays have been changed, and are now proportional to -log(p-value). The class frequencies (in functional analysis results) and the frequency ratios (in promoter analysis results) are now written on top of the columns. 8) Visualizations which include multiple histogram views have been changed so that each histogram has its own independent y-axis range, according to the hight of the columns. 9) A bug in the dendrogram visualization of hierarchical clustering has been fixed, so that the length of the branches is now proportional to the distance between the nodes. The visualization has been improved and it now contains a distance scale and a tool-tip which indicates the distance value at each point in the tree. the distance values range between 0 and 1 and are calculated as follows: distance = (- + 1)/2 10) Message boxes have been added and error messages have been improved. 11) Dialog boxes have been reorganized. 12) An option for exporting display into eps (post-script) format has been added. 13) A bug which caused problems when running SAMBA on Linux has been resolved.