Old mothur vs new mothur

Theory behind mothur
1

I’m analyzing a dataset published in 2011 and it analyzed it data using mothur, but before shhh.flows was implemented. Therefore, it seems to me that many of the sequences included in the study would be denoised by today’s version of mothur in shhh.flows. If I were to process the raw data in today’s mothur, I would get very different results.

My question is this: is this original study’s data rendered invalid if I use the shhh.flows and the new mothur to get better results? Are these new results really better results if I am throwing out information that is used in the study?

I know it may be a vague question but I’m just curious.
Thanks

2

It really depends on how things were analyzed in 2011. If the quality filtering were at all stringent, then it shouldn’t make much of a difference - except that with shhh.flows you would get more and longer reads. It’s hard to say what the effect would be short of doing the experiment.

Pat

3

My question is this: is this original study’s data rendered invalid if I use the shhh.flows and the new mothur to get better results? Are these new results really better results if I am throwing out information that is used in the study?

4

invalid might be a bit strong, but the new, denoised data are certainly better.