Gene Expression Analysis: technical aspects and methodologies
Introduction
Not so long ago, studying gene expression was limited to isolating mRNA, separating on a gel and transfer on a nylon membrane (the good old Northern Blot) and finishing off with hybridization with a radioactive probe specific to your gene of interest. This approach had many limitations:
- You needed a lot of good quality mRNA;
- To probe for multiple genes, you had to strip and re-hybridize, a procedure that could only be done a few times (3 to 4, with some luck…);
- Few samples could simultaneously be studied;
- It was no easy task to quantify gene expression levels.
- And of course, you needed to have a pretty good idea of which gene to use as probe!
Slide-based hybridization techniques that started to show up at the beginning of the 1990's changed all this. By inverting the process, using the mRNA as a pool of probes rushing to hybridize themselves on a collection of known gene sequences attached to a solid substrate, you now got access to the whole picture of the mRNAs present in a given sample. Because arrays are always built the same way in an industrial process (assuring, in theory, stability and quality), you could use as many samples as you wanted, the limit being the financial resources of your lab 
. Since the information that is generated from an array is digital(essentially, emitted light intensity), it is now possible to use software tools to analyze the results at a genome-scale resolution.
Plateforms
- Techniques using sequencing
- SAGE(Serial Analysis of Gene Expression)
- mRNA-SEQ
- Techniques using mechanical probe spotting
- Oligo/cDNA spotted arrays
- Techniques using in situ synthesis
- More to come…
Experimental Design
- More to come…
Analyzing your gene expression data
Introduction
To analyze gene expression data, you can use two different ways:
- Using tools that provide a Graphical User Interface (GUI), you analyze your data through the use of menus with options that can walk you through the analysis itself;
- Using the command line interface (CLI) with tools like R, an approach that requires not only some knowledge of the methods that you want to use but also the learning of a complex system of commands and parameters.
With our experience gathered in teaching the topic, we have observed that there is a need for both. The GUI road allows for a rapid understanding of the principles involved, it can get hard to use if you have a large dataset. The CLI methods on the other end are more flexible because you are free to do whatever you want on large datasets but the learning curve can be steep. This is why we have developed a tutorial serie for each approach.
If you are not using an Impilo server, you need to have the following installed on your machine:
- From the command line
- With a GUI
The steps involved in gene expression analysis
- More to come…
Gene Expression Public Databases
- More to come…