The DeCyder™ image analysis software has been uniquely developed for analysis of 2D-DIGE images. It has novel algorithms for co-detection of three multiplexed gel images. It also has high sensitivity and a wide dynamic range (5 orders of magnitude). DeCyder™ image analysis software performs the following functions: spot detection, background subtraction, in-gel normalization, artifact removal, gel-to-gel matching followed by statistical analysis. This product has been automated with respect to most of these parameters; however, some manual confirmation is required.
DeCyder™ spot detection is based on initial detection of the protein spots in the Cy2 image of pooled sample (internal standard) followed by application of similar spot boundaries to the remaining images (Cy3 and Cy5) within each gel. The spot quantification is performed by automatic normalization of spot volumes from the experimental groups (usually Cy3 and Cy5) against the internal standard (Cy2). The generation of identical spot detection patterns on all the images from a gel is made possible by the powerful algorithm. The internal standard as well as the multiplexing capability of 2D-DIGE virtually eliminates gel-to-gel variation, thereby making the image analysis by DeCyder ™ more robust. The 3-D spot viewer which has been incorporated into the software program is very helpful for the purposes of manual land-marking and confirming detection of true spots, as well as spot-matching across the entire subset of gels in an experiment.
The DeCyder™ image analysis software includes Difference In-gel Analysis (DIA) and DeCyder Biological Variation Analysis (BVA) modules. DIA is generally used for pair-wise comparisons of samples within a single gel. It is also useful for preliminary analysis between different phenotypes and generating validity of normal distribution for detected spots across the gel. The BVA performs quantification of protein expression across multiple gels. Primarily, the BVA provides an advanced level of analysis for discovery of differentially expressed proteins between experimental groups consisting of many samples. It also allows the application of statistical tests. The tests routinely employed are the Student’s t-test as well as one-way and two-way ANOVA for generating accurate quantification results for differentially expressed protein spots between various experimental groups.
The XML Toolbox facilitates automatic generation of user specified reports. The protein spots of interest (differentially/non-differentially expressed) can be specified by the user to generate a pick list for further characterization by mass spectrometry. We routinely use human post-mortem as well as non-human primate brain tissue to study complex mammalian behavior patterns in various neuropsychiatric disorders. The expression analyses of different phenotypes are studied by 2D-DIGE for which DeCyder™ plays a fundamental role in image analysis.
The spot filtering parameters usually have to be optimized according to the distribution of the proteins spots and their intensities produced by a particular 2D gel set-up. We have optimized parameters for spot filtering for the brain samples we run routinely with minor variations for spot detection in the DIA mode; slope > 1.0, area < 350, peak height < 350 and volume < 100,000. The detected spots are authenticated manually for all the gels. This is time consuming if your spot filtering parameters are not optimized to exclude spot artifacts and include true spots. Spot maps from all the gels from an experiment are first matched by manual landmarks and then in automatic mode by DeCyder™ BVA. The land-marking becomes very important if the protein spots across all the 2D gels in an experiment have not resolved uniformly. Eventually a t-test is useful for identification of statistical significant differences in expression levels of thousands of protein spots in various groups.
Dr. Nilesh S. Tannu
Research Associate
Wake Forest University, School of Medicine
Physiology and Pharmacology