For open source software, just about anything you could need is available in R (http://cran.r-project.org) & BioConductor (http://www.bioconductor.org/).  It will take you a bit of time to learn, but honestly, your quest should not be for the easiest to use software, but instead should be focused on finding the best and most appropriate analytical tool(s) for your data.

The packages in BioConductor typically have a vignette in PDF format and will download with example data to work through to learn the commands for the package.  There are also online tutorials you can find with a web search, depending on just what kind of data, and what kind of analyses you wish to do.

I have used Dchip a simple software to analyze micro array data, but one must have little knowledge about Microsoft Excel data processing.

I wish to know if you would like to analyze micro array data that was obtained through your experimentation? or If u wish to analyze already available datasets in NCBI?

For open source software, just about anything you could need is available in R (http://cran.r-project.org) & BioConductor (http://www.bioconductor.org/).  It will take you a bit of time to learn, but honestly, your quest should not be for the easiest to use software, but instead should be focused on finding the best and most appropriate analytical tool(s) for your data.

The packages in BioConductor typically have a vignette in PDF format and will download with example data to work through to learn the commands for the package.  There are also online tutorials you can find with a web search, depending on just what kind of data, and what kind of analyses you wish to do.

For Affymetrix arrays, the Expression Console + Transcriptiome Analysis Console are very helpful and user-friendly free tools. For Affymetrix and Illumina chips the most simple tool to use is web-based  www.arrayanalysis.org - it does not require any SW installation, just uploading raw data files (.CEL), and it performs normalization, quality control and unsupervised class discovery analysis. 

I am using Multiple Array Viewer in MAC, if that helps. Some very basic preliminary analyses, such as K-means clustering, can be easily done.

you can use geWorkbench 2.5.1 from NCI. You can download from:

https://wiki.nci.nih.gov/display/geWorkbench/geWorkbench

as commented before, R and bioconductor can provide several packages. However, if you are looking for a web based friendly interface, Babelomics (http://babelomics.bioinfo.cipf.es/) is a complete suite that allows basic preprocessing, clustering and functional analysis (which includes, among others, gene set enrichment analysis and PPI network construction and analysis).

If you want something that doesn't need programming skills, MultiExperiment Viewer from TIGR is worth a try: http://www.tm4.org/mev.html

Simple to download, install and use. It has useful tools like PCA, clustering and statistical tests - BUT your expression data will need to be normalised first.

I am trying to use AltAnalyzis. Does anyone has an experience and/or a feedback on this one?

AltAnalyze is a pretty simplistic approach to microarray analyses with limited options (last I ever looked at it, the only statistical test possible was one-way ANOVA).  Depending on your experiment, it might be alright or it might not (how complex is your experimental design?).

Seriously, I just think you are approaching this entirely in the wrong way.  You need to identify exactly WHAT analyses you want to do (or need to do) given your experimental design and original hypothesis, and then seek out the appropriate tools to do those analyses.  If that means learning some more complex software, then so be it, if the more complex software is the most appropriate tool for your data.  You learn whatever statistics and software you have to in order to properly analyze your data.

If you have a multifactoral experiment where you need to look at several contrasts simultaneously, then no, AltAnalyze is not going to be appropriate - something like fitting a GLM in LIMMA with empirical Bayesian stats might be far more appropriate.  If you have a very small data set with limited replicates, then an ANOVA would not be my first choice of statistical test to maximize the potential findings in such a data set.

But you don't pick the tool and then force the analysis to fit that tool.  You pick the analysis first, based on the most appropriate treatments and tests for your data, and then you find a tool that implements that analysis for you.  Given the abundance of open source tools for microarray data, a free tool is almost certainly available, but which one depends on your experiment and data.

AltAnalyze actually supports a large number of statistical methods inclusibeg LIMMA based empirical moderated t-test and Benjamini-Hochberg adjustment. Besides being the easiest to use free application that I am aware of, it includes dozens of automated computational methods such as customizable heatmap visualization, comprehensive splicing and isoform function analyses, supports RNA-Seq, conventional and splicing arrays, PCA, identifier translation, file merging, Venn diagrams, QC plots, Wikipathway visualization and pathway/geneset/TF/miR/ontology enrichment analyses, network analysis and visualization, splicing graph visualization, marker identification, sample classification and accurate cell-type prediction analyses. One feature that is particularly useful is the advanced clustering and heatmap visualization options, which include the option to filter large datasets, use hopach clustering, perform correlation based queries, filter on many different ontologies and gene sets and selectively view specific genes of interest. We have recently added novel methods for single-cell seq analysis. There is extensive documentation, active development and a large user community. Many of these methods are still being improved so please provide feedback.

I can suggest using ExAtlas http://lgsun.grc.nia.nih.gov/exatlas/ which is on-line software that automatically retrieves data from GEO database together with NCBI-updated probe annotations. No need to installation, just upload your data to the server. In contrast to other software, it compares multi-component data sets and generates results for all combinations (e.g., all gene expression profiles vs. all GO annotations). Main functions are: (1) standard meta-analysis (fixed and random effects, z-score, Fisher's method; analysis of (2) global correlations between gene expression data sets; (3) gene set enrichment; (4) gene set overlap; (5) gene associatio; (6) statistical analysis of gene expression (ANOVA, PCA, FDR). Results are presented graphically as heatmaps, bar-charts, or 3-D images (PCA). Several most popular public data sets (e.g., GNF, Gene Ontology, KEGG, GAD phenotypes, protein interactions) are pre-loaded and can be used for functional annotations.

NetworkAnalyst is user friendly and freely available gene expression analysis tool http://www.networkanalyst.ca/faces/docs/FaqView.xhtml

Hello Christian. Using the Bioconductor Package with the R Program is a really great way to read Microarray Gene Expression Data, conduct multiple analyses, and create great 3D data visualizations (Principal Component Analysis, Contrast Heatmaps, MA Plots, Cluster Dendrograms). We are also developing a software process that can find frequently expressed genes when combining multiple microarray datasets, keeping track of where the expressed genes occur, taking account of how the microarray data is recorded (ex: non-traceable, direct assay analysis, etc), and creating databases and models based on Gene Ontology (Biological Process, Cellular Components, Molecular Functions). Our software processes and algorithms are very easy and in fact we have created a database and software process for a technology transfer office (they are still using it) that helped them secure patents and break their financial records this past year. If anyone has questions, feel free to email me at [email protected]. Thank you and best of luck with your professional endeavors.

you can use Image J software also for analysis of gene expression.