Analyzing Multifactorial RNA-Seq Experiments with DiCoExpress

DiCoExpress provides a complete aoristic analysis from quality control to co-expression. It performs differential analysis based on contrasts inside the generalized linear model. Moreover, it can also perform an enrichment analysis on the list of differentially expressed genes and the co-expressed gene clusters.

The main advantage of DiCoExpress is that it can be used by people just like me, without any particular knowledge in statistics or air programming. It truly helps a non-specialist user to write the contrast necessary for differential gene expression analysis. It also provides graphical outputs illustrating the results ready for publication.

DiCoExpress is not a plan dedicated tool. It can be used for any organism as long as the experimental design is complete with up to two biological factors. Moreover, a membrane's design with a unequal number of replicates between condition is also possible.

A beginner should have preliminary knowledge in R.You should know how to use a function and identify required and optional arguments. Then the critical step is to correctly provide the files containing the and the experimental design. To begin, open the R studio session.

Set the directory to template scripts and open the DiCoExpress tutorial dot R script. Load the DiCoExpress functions in the R session. Then load data files in the R session and split the object data files into several objects for manipulating the files easily.

Next, select a strategy among NB conditions or NB replicates and a threshold to filter low express genes. Specify group colors and select a normalization method. Then perform the quality control.

If data are paired according to the replicate factor state replicates as true, otherwise state as false. Assign the interaction as true to consider an interaction between the two biological factors. Otherwise assign false, then specify the statistical model and define the threshold of the false discovery rate.

Perform the differential analysis followed by fixing a threshold for the enrichment analysis and performing the enrichment analysis of differentially expressed gene lists. Select the DEG lists to be compared. Provide a name for the list comparison and use the same name for the directory where the output files will be saved.

Set the parameter operation to union or intersection for specifying the action to be done on the DEG lists, and compare the lists. Carry out a co-expression analysis followed by conducting the enrichment analysis of the co-expression clusters. And at last, generate two log files containing all the necessary information to reproduce the analysis.

The total normalized counts per sample should be similar when comparing both intra and inter conditions. The normalized gene expression counts exhibited similar median and variance, both in intra and inter conditions. For identifying the potential underlying data structures, PCA plots were generated.

A clear distinction was observed between the treatments and clustering was absent, indicating a good quality data set. The raw pvalue histograms were plotted to assess the quality of the modeling. The distribution of raw pvalues was uniform, with a peak at the left end side of the distribution, as expected.

The absence of a peak at the right end side indicates that the statistical modeling seems correct. The expression profile of Gene CIG62301.1, in every genotype and condition, was plotted. As well as the number of up and down differentially expressed genes, were also plotted for every tested contrast.

The co-expression analysis was performed on the union of five DEG lists. Identified by contrast, looking for treatment response variation between genotype one or two against others. The co-expressed genes for every identified cluster were printed in individual text files and the expression profile of genes was plotted.

With DiCoExpress biologists will get gene expression analysis that are statistically sound. The next step is make biological sense, out of these results.