When there are multiple runs within a single experiment (in RNA-seq analysis)

Question

[wyt1995]

Hello, I'm a student who hasn't done much analysis before. I would appreciate it if you could understand even if my question is somewhat basic. I wanted to perform quality control, adapter trimming, mapping, and quantification on the raw data from a dataset available on GEO. I downloaded all the data from the SRA of SRP074349 within the GSE81089 dataset. However, I noticed that there are multiple SRRs (runs) for several SRXs (experiments).

In this case, should I choose only one, merge them into one, or filter them using FastQC or similar tools? I'm curious about the best approach.

Answer 1

[GenoMax]

This is a large project (670 samples) and you can get a better look at the sample metadata using SRA Run selector: https://www.ncbi.nlm.nih.gov/Traces/study/?query_key=1&WebEnv=MCID_6464a81284083773086ece4b&o=sample_name_s%3Ad%3Bacc_s%3Aa

If you sort the table by "Sample Name" you will see that there are some samples with multiple SRA numbers. These are more than likely technical replicates that can be merged into one at some point in data processing.

But for others you will want to check the metadata to see if metadata distinguishes them e.g. tumor/normal from same patient.

[wyt1995]

When I checked other answers, I also saw that when there are multiple runs (SRR) within a single experiment (SRX), it was mentioned to merge them. Could you please provide me with a link to a resource where I can learn more about this merging process?