Help to understand some terms

Question

[yuvaltechnion]

I'm new to the field of bioinformatics (and I'm not from the field of biology), and I don't understand some terms in gene expression analysis:
I read about RNA-seq that it's important to use some replicates. What is the meaning of replicates here? I know the process of dna replication, but it does not seem useful to use copies of the same DNA.
And what exactly is outlier in RNA-seq analysis? I read that the more replicates we have the easier it is to detect outliers. Can you explain it to me please?

Thanks.

Answer 1

[Lars Juhl Jensen]

The words replicates and outliers are terms from statistics. Wikipedia gives a very good explanation of both:

• https://en.wikipedia.org/wiki/Replication_(statistics)
• https://en.wikipedia.org/wiki/Outlier

Long story short, replicates means that you do the experiment more than once and outliers typically represent experimental errors. Obviously, if you repeat an experiment, it makes it a lot easier to identify erroneous measurements.

[WouterDeCoster]

To add a simple example to this:
If you do 3 measurements and get values 5 9 and 6 you don't know if those values belong to the same distribution, or maybe that 9 is an outlier.
If you do 12 more measurements and get values 5 9 6 4 5 6 5 6 6 4 6 6 4 6 4 it does become clear that 9 could be an outlier.

Both for this artefact as for true biological variability: more replicates give you more power to make discoveries