# is there differece with merge feature-table and not merge

**URL:** https://forum.qiime2.org/t/is-there-differece-with-merge-feature-table-and-not-merge/21802
**Category:** General Discussion
**Created:** [January 6, 2022, 8:17am UTC](https://forum.qiime2.org/t/is-there-differece-with-merge-feature-table-and-not-merge/21802 "2022-01-06T08:17:34Z")
**Posts on this page:** 1
**Showing post:** 4

<div class="post-metadata">

### Author: ![colinbrislawn](https://forum.qiime2.org/user_avatar/forum.qiime2.org/colinbrislawn/32/6221_2.png) [@colinbrislawn](https://forum.qiime2.org/u/colinbrislawn)
#### Post date: [January 10, 2022, 2:18pm UTC](https://forum.qiime2.org/t/is-there-differece-with-merge-feature-table-and-not-merge/21802/4 "2022-01-10T14:18:10Z")

</div>

Ah ok!

DADA2 builds an error model for each sequencing run, so it's best to run it separately on seperately runs. See [this form post](https://forum.qiime2.org/t/dada2-samples-from-multiple-miseq-sequencing-runs/1311/4) and [this part](https://benjjneb.github.io/dada2/bigdata.html) of the dada2 tutorial.

For samples on the same run, dada2 should produce **equivalent** \* results wether run on one sample or all!

Let us know if you have any other questions! 🙋‍♂️

EDIT:  
\*But probably not **identical** , because

> [@Mehrbod\_Estaki](#):
>
> the training reads will differ between runs and so you'd get small minor differences

---

_[View the full topic](https://forum.qiime2.org/t/is-there-differece-with-merge-feature-table-and-not-merge/21802)._