# Suggestions for using nifH ARB database for taxonomy assignment in QIIME2

**URL:** https://forum.qiime2.org/t/suggestions-for-using-nifh-arb-database-for-taxonomy-assignment-in-qiime2/14694
**Category:** General Discussion
**Tags:** taxonomy
**Created:** [April 29, 2020, 1:52pm UTC](https://forum.qiime2.org/t/suggestions-for-using-nifh-arb-database-for-taxonomy-assignment-in-qiime2/14694 "2020-04-29T13:52:38Z")
**Posts on this page:** 1
**Showing post:** 16

<div class="post-metadata">

### Author: ![SoilRotifer](https://forum.qiime2.org/user_avatar/forum.qiime2.org/soilrotifer/32/21071_2.png) [@SoilRotifer](https://forum.qiime2.org/u/SoilRotifer)
#### Post date: [December 5, 2022, 9:59pm UTC](https://forum.qiime2.org/t/suggestions-for-using-nifh-arb-database-for-taxonomy-assignment-in-qiime2/14694/16 "2022-12-05T21:59:02Z")

</div>

Hi @Mitra_Ghotbi,

I exported your `fungene_8.1_nifH_ref_unaligned_nucleotide_seqs.qza` as a FASTA file. The issue is that the FASTA headers (a few output below):

```auto
>821566130location=complement(110918..111811),organism=RhizobiumphaseoliCh24-10,definition=nitrogenasereductaseNifH

```

are different from your taxonomy headers:

```auto
821566130 k __Bacteria; p__ Proteobacteria; c __Alphaproteobacteria; o__ Rhizobiales; f __Rhizobiaceae; g__ Rhizobium; s__phaseoli Ch24-10

```

I was going to suggest that you either need to remove everything after the `>821566130` in the FASTA header, or simply insert a space after `821566130`. However, this won't work as there appears to be multiple sequences with the same ID, _i.e._ there are three sequence entries for `821566130`:

```auto
>821566130location=complement(110918..111811),organism=RhizobiumphaseoliCh24-10,definition=nitrogenasereductaseNifH
>821566130location=68723..69616,organism=RhizobiumphaseoliCh24-10,definition=nitrogenasereductaseNifH
>821566130location=complement(16224..17117),organism=RhizobiumphaseoliCh24-10,definition=nitrogenasereductase

```

Each sequence ID must be unique and correspond to a single unique taxonomy ID. For more details on appropriate IDs, see [here](https://docs.qiime2.org/2022.8/tutorials/metadata/#identifier-column).

If you'd like to keep all the sequences, then I'd arbitrarily increment the ID like so:

> 821566130  
> 821566130.1  
> 821566130.2

or append the location information like this (I am adding 'c' to denote compliment). This is similar to how SILVA, and other databases, handle multiple gene copies from the same organism:

> 821566130.c110918  
> 821566130.68723  
> 821566130.c16224

Then make sure the IDs in the taxonomy file match those in the sequence file. Then you should be good to go.

_Alternatively you can try [RESCRIPt](https://github.com/bokulich-lab/RESCRIPt), to make your own nifH reference database. You can look through this [tutorial](https://forum.qiime2.org/t/using-rescripts-extract-seq-segments-to-extract-reference-sequences-without-pcr-primer-pairs/23618) ._

---

_[View the full topic](https://forum.qiime2.org/t/suggestions-for-using-nifh-arb-database-for-taxonomy-assignment-in-qiime2/14694)._