# how reliable are my taxonomic classifications if I did not identify species?

**URL:** https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696
**Category:** General Discussion
**Created:** [December 5, 2020, 12:57am UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696 "2020-12-05T00:57:08Z")
**Posts on this page:** 8
**Page:** 1

<div class="post-metadata">

### Author: ![Nisha](https://forum.qiime2.org/user_avatar/forum.qiime2.org/nisha/32/11057_2.png) [@Nisha](https://forum.qiime2.org/u/Nisha)
#### Post date: [December 5, 2020, 12:57am UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696/1 "2020-12-05T00:57:08Z")

</div>

Thanks

Also I had a query.  
There are 25000 OTUs identified by Qiime2 in my samples, but after I got annotations only for 9000 at species level, in that too, species names are unassigned. How reliable these results are?

---

<div class="post-metadata">

### Author: ![Nicholas\_Bokulich](https://forum.qiime2.org/user_avatar/forum.qiime2.org/nicholas_bokulich/32/19937_2.png) [@Nicholas\_Bokulich](https://forum.qiime2.org/u/Nicholas_Bokulich)
#### Post date: [December 5, 2020, 7:23am UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696/2 "2020-12-05T07:23:13Z")

</div>

> [@Nisha](#):
>
> but after I got annotations only for 9000 at species level, in that too, species names are unassigned. How reliable these results are?

Sounds much more reliable than getting species identifications for everything! I recommend checking out the literature to see some extensive benchmarks on how read length, marker gene target, and subdomain targets impact taxonomic resolution and accuracy. See these articles for starters:

> **[Optimizing taxonomic classification of marker-gene amplicon sequences with...](https://microbiomejournal.biomedcentral.com/articles/10.1186/s40168-018-0470-z)**
>
> Background Taxonomic classification of marker-gene sequences is an important step in microbiome analysis. Results We present q2-feature-classifier ( https://github.com/qiime2/q2-feature-classifier ), a QIIME 2 plugin containing several novel...

[https://aem.asm.org/content/73/16/5261](https://aem.asm.org/content/73/16/5261)

---

<div class="post-metadata">

### Author: ![Nisha](https://forum.qiime2.org/user_avatar/forum.qiime2.org/nisha/32/11057_2.png) [@Nisha](https://forum.qiime2.org/u/Nisha)
#### Post date: [December 7, 2020, 2:45pm UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696/3 "2020-12-07T14:45:54Z")

</div>

but I am worried about the genera or species, which are most abundant and still unassigned.

---

<div class="post-metadata">

### Author: ![Nicholas\_Bokulich](https://forum.qiime2.org/user_avatar/forum.qiime2.org/nicholas_bokulich/32/19937_2.png) [@Nicholas\_Bokulich](https://forum.qiime2.org/u/Nicholas_Bokulich)
#### Post date: [December 7, 2020, 3:38pm UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696/4 "2020-12-07T15:38:26Z")

</div>

If the sequence is abundant but has no species label, that is typical for most short read sequences of 16S rRNA gene variable regions. Unassigned genus is still not uncommon for, e.g., V4.

If the sequence is abundant and is not even assigned at phylum level, it is probably non-target DNA... this is common, e.g., to amplify plant chloroplast DNA or other host DNA from some sample types.

---

<div class="post-metadata">

### Author: ![Nisha](https://forum.qiime2.org/user_avatar/forum.qiime2.org/nisha/32/11057_2.png) [@Nisha](https://forum.qiime2.org/u/Nisha)
#### Post date: [December 7, 2020, 6:39pm UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696/5 "2020-12-07T18:39:01Z")

</div>

I have 16S rRNA PE data for V1-V3 region  
What should I do at this stage?

---

<div class="post-metadata">

### Author: ![Nicholas\_Bokulich](https://forum.qiime2.org/user_avatar/forum.qiime2.org/nicholas_bokulich/32/19937_2.png) [@Nicholas\_Bokulich](https://forum.qiime2.org/u/Nicholas_Bokulich)
#### Post date: [December 9, 2020, 7:24am UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696/6 "2020-12-09T07:24:47Z")

</div>

> [@Nisha](#):
>
> I have 16S rRNA PE data for V1-V3 region

Same issue. Species can be unreliable due to lack of species-level resolution in this region.

> [@Nisha](#):
>
> What should I do at this stage?

See my post above, but assuming you have family and genus level classifications, I'd say just move on. You could try out q2-clawback to help guide species-level classifications:

> [@Using q2-clawback to assemble taxonomic weights](https://forum.qiime2.org/t/using-q2-clawback-to-assemble-taxonomic-weights/5859):
>
> Note: This guide assumes you have installed QIIME 2 using one of the procedures in the [install documents](https://docs.qiime2.org/2019.1/install/). This tutorial extends the q2-feature-classifier [tutorial](https://docs.qiime2.org/2019.1/tutorials/feature-classifier/) to show how you can improve taxonomic classification accuracy by assembling [ecologically informed taxonomic weights](https://doi.org/10.1038/s41467-019-12669-6) for your samples using q2-clawback. This approach is described in this article: [https://doi.org/10.1038/s41467-019-12669-6](https://doi.org/10.1038/s41467-019-12669-6) We will retrain the naive Bayes classifier from the q2-feature-classifier to classify the ski…

---

<div class="post-metadata">

### Author: ![Nisha](https://forum.qiime2.org/user_avatar/forum.qiime2.org/nisha/32/11057_2.png) [@Nisha](https://forum.qiime2.org/u/Nisha)
#### Post date: [December 12, 2020, 9:06am UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696/7 "2020-12-12T09:06:29Z")

</div>

I have reduced my analysis till genus level..but here also I have found a lot of OTUs whose genera are unavailable.

These are one among most prevalent and most abundant genera.

---

<div class="post-metadata">

### Author: ![Nicholas\_Bokulich](https://forum.qiime2.org/user_avatar/forum.qiime2.org/nicholas_bokulich/32/19937_2.png) [@Nicholas\_Bokulich](https://forum.qiime2.org/u/Nicholas_Bokulich)
#### Post date: [December 12, 2020, 5:55pm UTC](https://forum.qiime2.org/t/how-reliable-are-my-taxonomic-classifications-if-i-did-not-identify-species/17696/8 "2020-12-12T17:55:05Z")

</div>

> [@Nisha](#):
>
> I have reduced my analysis till genus level…but here also I have found a lot of OTUs whose genera are unavailable.

This is really a common issue with 16S rRNA gene sequences, I must say.

> [@Nisha](#):
>
> These are one among most prevalent and most abundant genera.

I recommend reading the forum archives more to see how others have dealt with this issue (usually, by doing nothing, just accepting that short 16S reads have limited taxonomic resolution). The FAQs are a great start for various taxonomy-related topics:

> [@Frequent Questions and "Best of the QIIME 2 Forum"](https://forum.qiime2.org/t/frequent-questions-and-best-of-the-qiime-2-forum/13608):
>
> The following is a list of frequently asked questions on the forum, with answers, to serve as a point of reference for forum users to quickly find answers, or for forum users and moderators to use as a reference when helping others. Note: some of these listings are collapsed lists. If you see an arrow next to a title, click the title to expand the list of topic titles related to that subject. arrow_forwardFAQs [What is a "Feature" exactly? - #3 by Nicholas\_Bokulich](https://forum.qiime2.org/t/what-is-a-feature-exactly/2201/3)[MemoryError with q2-f…](https://forum.qiime2.org/t/feature-classifiers-error/11946/2)

Good luck!
