# Custom databases usage

**URL:** <https://forum.biobakery.org/t/custom-databases-usage/1737>\
**Category:** HUMAnN\
**Created:** [March 2, 2021, 2:35pm UTC](https://forum.biobakery.org/t/custom-databases-usage/1737 "2021-03-02T14:35:07Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Sergey\_Petrov](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/sergey_petrov/32/650_2.png) [@Sergey\_Petrov](https://forum.biobakery.org/u/Sergey_Petrov)\
**Post date:** [March 2, 2021, 2:35pm UTC](https://forum.biobakery.org/t/custom-databases-usage/1737/1 "2021-03-02T14:35:07Z")

</div>

Hello

I’ve created custom no uniref-based protein database with arbitrary sequence headers. How can i link it to custom pathway database?

For example:

protein db:

> ref\_prot1  
> \<…\>  
> ref\_prot2  
> \<…\>

Pathway db, file 1:  
Rxn-1 ref\_prot1  
Rxn-2 ref\_prot2

Pathway db, file 2:  
PWS-TEST Rxn-1 Rxn-2  
(in accordance with [GitHub - biobakery/humann: HUMAnN is the next generation of HUMAnN 1.0 (HMP Unified Metabolic Analysis Network).](https://github.com/biobakery/humann#custom-pathways-database))

Diamond generates output tsv file with alignment against custom db but humann reports that total gene families count is zero. What am i doing wrong?

---

<div class="post-metadata">

**Author:** ![franzosa](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/franzosa/32/3511_2.png) [@franzosa](https://forum.biobakery.org/u/franzosa)\
**Post date:** [March 11, 2021, 1:58pm UTC](https://forum.biobakery.org/t/custom-databases-usage/1737/2 "2021-03-11T13:58:30Z")

</div>

Are you testing on a very small metagenome? It’s possible that none of your custom proteins are being covered at 50% of sites, in which case HUMAnN will conservatively not report them. You can lower this threshold on the command-line (translated subject coverage threshold) such that any valid read-hit will add its protein to the output file.

---

<div class="post-metadata">

**Author:** ![Sergey\_Petrov](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/sergey_petrov/32/650_2.png) [@Sergey\_Petrov](https://forum.biobakery.org/u/Sergey_Petrov)\
**Post date:** [March 11, 2021, 4:39pm UTC](https://forum.biobakery.org/t/custom-databases-usage/1737/3 "2021-03-11T16:39:14Z")

</div>

I’ve created nucleotide dataset from 11321 sequences, where each sequence corresponds to some protein. Than i’ve translated them into aminoacid form and created diamond database from gained file.  
Then i’ve launched Humann with source fasta file as input data.  
The file diamond\_aligned.tsv which stores aligning against diamond database shows me that each nucleotide sequence was aligned agaisnt itself in amino form(just as expected). So, each protein in database is full covered by 1 read.

---

<div class="post-metadata">

**Author:** ![Kang\_Da](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/kang_da/32/2492_2.png) [@Kang\_Da](https://forum.biobakery.org/u/Kang_Da)\
**Post date:** [October 21, 2023, 5:59am UTC](https://forum.biobakery.org/t/custom-databases-usage/1737/4 "2023-10-21T05:59:03Z")

</div>

Hi, did you solve the problem? I met the same question, humann3 cannot align against my own protein database, but I can do it use only the diamond.

---

<div class="post-metadata">

**Author:** ![franzosa](https://yyz1.discourse-cdn.com/flex027/user_avatar/forum.biobakery.org/franzosa/32/3511_2.png) [@franzosa](https://forum.biobakery.org/u/franzosa)\
**Post date:** [October 26, 2023, 8:21pm UTC](https://forum.biobakery.org/t/custom-databases-usage/1737/5 "2023-10-26T20:21:39Z")

</div>

How have you formatted your custom database? Are you seeing an error, or just a lack of meaningful output?
