You need to do a bit of research first in order to obtain good results. It all depends on tissue, species, treatment, age etc. Read a paper on it here http://www.sciencedirect.com/science/article/pii/S0006291X03025646 and you should be familiar with MiQe guidelines (http://www.ncbi.nlm.nih.gov/pubmed/19246619) in order to design good experiments. Not all the guidelines are needed in all cases, but as many as possible will increase the quality of the experiment.
You need to do a bit of research first in order to obtain good results. It all depends on tissue, species, treatment, age etc. Read a paper on it here http://www.sciencedirect.com/science/article/pii/S0006291X03025646 and you should be familiar with MiQe guidelines (http://www.ncbi.nlm.nih.gov/pubmed/19246619) in order to design good experiments. Not all the guidelines are needed in all cases, but as many as possible will increase the quality of the experiment.
I agree with Dipanjan Chanda about the 18s. GAPDH and Actin can be tricky if you are comparing between different tissues or cell types. However, remember to dilute your sample, since the 18s RNA is much more concentrated than mRNAs. Also bear in mind that you will not be able to amplify 18s if you performed your retrotranscription using oligo-dT primers. You need to have a cDNA pool originated from total RNA, like the one you get when using random primers in your RT.
Internal Controls (or Internal Positive Controls) would depend on the species you are working with (as Anja Paschold said); however, it would also depend on the type of assay you are doing.
In general, if you are looking at gene expression studies, then an endogenous reference gene (such as GAPDH, beta-actin or 18S rRNA as mentioned by Priya Arora) would be used to calculate the fold change (up-regulation or down-regulation) of a particular gene of interest (for test versus control) by using 2 raised to -delta delta cT method. in this case it is not an 'internal control' in the truest sense.
Most internal positive controls are used to validate whether your extraction procedure or PCR reaction is working fine or not. In such cases, especially when using TaqMan chemistry, another set of primers and probe with a different reporter dye (than that for your actual target gene) is used in a multiplex reaction. The choice of reporter dyes of course would depend on the type of real time instrument you have in your lab.
in these cases too the same reference genes can be used as internal controls, but you may also separately add a gene (for example in the form of a plasmid) from a non-related species directly into your sample prior to extraction to validate your entire assay starting from extraction followed by real time PCR.
18 S rRNA is definitely better than Actin and GADPH, the reason being that it shows comparatively less variance in expression across a variety of treatment conditions.
definitely it depends on the cell type and treatment but in my experience RPLP0 seems to be quite good and even better than 18S rRNA. Other useful are POLR2A or IPO8. It is always better to run more housekeeping genes and compare the result so as you can see which one seem to be more suitable for your particular purpose
Also, do not forget that it depends upon the fact if you are using random hexamer primed cDNA or oligo dT primed cDNA. 18S rRNA will not work if you use the latter method for producing cDNA.
I'd recommend RPL13A as a reference gene for q-RT-PCR. It is abundant and has been shown not to vary too much in expression level in different types of sample types. However, the best way may be to use a combination of two reliable housekeeping gene expression for internal normalization after a preliminary experiment to test the relative variance of the expression of these two internal controls across different sample types.
1) it depends on the species, tissue treatment etc. as above mentioned.. A very good read for choosing the right one is this... http://www.ncbi.nlm.nih.gov/pubmed/12184808
Thank you all of you for suggestion... Can we Design our own Internal Control for test validation..? Because during the DNA isolation, if any problem occurs and we r not able to getting housekeeping gene or volume is very less, in that condition our own internal Control will works....or not..?