dChip: Introduction and Installation

Memory requirement User forum and trouble-shooting

Running on Mac or Unix Source code and development

Automated analysis

Introduction

DNA-Chip Analyzer (dChip) (www.dchip.org) is a Windows software package for probe-level (e.g. Affymetrix platform) and high-level analysis of gene expression microarrays and SNP microarrays (Li and Wong 2001a, Lin et al. 2004). Gene expression or SNP data from various microarray platforms can also be analyzed by importing as external dataset. At the probe level, dChip can display and normalize the CEL files, and the model-based approach allows pooling information across multiple arrays and automatic probe selection to handle cross-hybridization and image contamination. High-level analysis in dChip includes comparing samples, hierarchical clustering, view expression and SNP data along chromosome, LOH and copy number analysis of SNP arrays, and linkage analysis. In these functions the gene information and sample information are correlated with the analysis results.

Download the dChip software (e.g. dchip2006.exe) and gene information files and store them into a local directory, and double-click the program icon to start dChip. This manual is written in a tutorial style. You may explore the functions that interest you by following the steps described in the manual. Paragraphs labeled as “[Analysis example]” may be skipped in the first reading. Please cite Li and Wong 2001a if dChip results are used in manuscripts, and cite Lin et al. 2004 if dChip SNP analysis functions are used.

Other free microarray analysis software include: Cluster/TreeView, BioConductor, BRB ArrayTools, Cytoscape, GenePattern, GenMAPP, GoSurfer, GSEA, SAM, TM4.

Memory requirement

dChip is a single executable program developed on Windows XP (Windows 2000 before 2006). It should run properly on Windows 2000/NT/XP, but some functions such as clustering picture zooming and image saving do not work properly on Windows 95/98. dChip is built as 32 bit program but it can run on 64 bit computer.

The limit of the number of arrays analyzed in dChip is 2000. However the practical limit is dependent on computer memory. Computer with 1 GB or more memory is preferred. Version 4/14/06+ is made to be “large address aware” so it can access 3-4 GB memory (MS information, information 2), and this link shows how to set the "/3GB" option on a Win 32 bit PC to use 3G memory. When analyzing several hundred arrays, "Analysis/Open group" and "Analysis/Chromosome" may report "out of memory" error. Uncheck "Open group/Options/Load probe data in memory" may help in such situations.

User forum and trouble-shooting

Please use the dChip user forum for dChip related correspondences such as usage questions and bug reports. Please include the following information if relevant:

• The complete analysis output of the entire dChip session, including dChip error messages and version dates. At the Analysis view window, use Control+A to select all texts and then Control+C or “Analysis/Copy”, or use “Analysis/ Save” to save output into a file.
• dChip exported data files or image files.
• Screenshots of the issue in question, obtained by pressing the PrintScreen key and pasting into the Paint software to save.
• Use “Tools/Options/Print Settings” to print out the setting information of the current analysis session.

Here are some quick solutions that you may try:

• Click the "Tools/Options/Reset Default" button to reset settings, or delete the existing “group_name”.ini files (in the same directory as dchip.exe) and then restart dChip.

· Use the latest Gene information files or dChip. If needed, copy these files and the CEL data files (but not the *.ini files containing dChip settings) into a new directory, and start dChip from here.

· Uncheck “Analysis/Open group/Options/Load probe data in memory” if a group contains hundreds of arrays or arrays with very high density (e.g. 500K SNP array.

• When opening a group, check “Open group/ignore existing DCP file” and “Open group/Other information/ignore existing cdf.bin file” to re-extract the CEL and CDF files.
• External data file, gene information file, sample information file may have been saved in the Excel format. Open files in Excel and use “File/Save As” to save in tab-delimited text format.
• If “out of memory” when clustering many genes, you can uncheck “Tools/Options/Clustering/Pre-calculate distance” before clustering.
• In the clustering or chromosome view, the part of figure is overlapped or not displayed: use Arrow keys, Control+Arrow keys and Shift+Arrow keys to adjust.
• Clustering image exporting problem.

Running on Mac or Unix

Virtual PC software enables running dChip on Max and Unix computers. For example, Virtual PC for Mac and Wine. Victoria Perreau has run dChip on an iBook, using the Connectix Virtual PC Windows 2000 software which costs about $225. When running on the iBook (466MHz, 320 RAM) it takes about 7 times longer to do the analysis than a PC (1000Mhz Pentium III , 256RAM), i.e. 6 CEL files are normalized in 8 minutes instead of 1. Matthias Klaften runs dChip2006 on a SuSE Linux 9.0 box (Kernel 2.4.21) with Wine 0.9.6. However, errors may occur when running in these platforms and we have not tested.

Source code and development

[Forum] dChip is written in Visual C++ 6.0/2005 and uses Windows-specific graphic functions. The latest source codes of dChip are available on request (Cheng Li). It can be used for developing new or customized functions for microarray analysis. It can also be compared to the 2002 command-line codes below to see what parts need to be changed to build on Unix. If you build a Unix or Linux version, we will appreciate it if you can share the codes or binaries or be linked from this page.

User-built dChip versions: Linux/MPI by Luca Corradi (Forum thread). (7/16/07: Start linking to forum threads from user's manual.)

A user is encouraged to use the standard dChip version for full functionalities. A dChip command-line version has been adapted (V1.2 Test, 10/11/02, dchip_2002_09.zip). If _WINDOWS is not defined, it builds a command line version in windows, or builds on Unix. It continuously executes the "Analysis/Open group", "Analysis/Normalize", "Analysis/Model-based expression" and "Tools/Export data/Expression value" steps. The result is a tab-delimited text file containing the model-based expression values, along with the "array summary file". The relevant parameters such as the directory containing CEL or DAT files (DATA_PATH), CDF file name (CDF_FILE) and working directory (WORKING_DIR) are specified in the configuration file "dchip.ini" located in the same directory as "dchip_console.exe". One may run the program once to create the "dchip.ini" file and then edit the file to specify parameters. In addition parameters for different groups can be specified by using command such as "dchip_console.exe group_name.ini". The model-based expressions calculated on different platforms may not be exactly the same. This is probably due to the platform-dependence of number type precision and numerical libraries. The value differences may further cause different detection of outliers by dChip.

(Updated 12/12/07)