meta-analysis using different sequencing platforms

General Discussion
,
1

Hi everyone,
I wanted to use data from studies using ion-torrent and Illumina platforms for a meta-analysis. Any thoughts on that, is it advisable, what are the things I should keep in mind if attempting anything like that?
Any thoughts will be appreciated.

Thanks,
Sudhir

2

Hi @sudhir19, meta-analysis like you're describing is possible but must be done very carefully and you should constantly be on the lookout for differences arising from the different sequencing platforms. And if sequencing run is perfectly correlated with an experimental variable of interest (e.g., if all control samples are on one run and all experimental samples are on the other run) deriving conclusions (rather than generating hypotheses) from the data is inadvisable.

There is a lot of discussion on the QIIME 2 Forum about meta-analysis. Please start by taking a look through the existing posts, and feel free to ask again if you have specific questions.

Good luck!

2 Likes