Installing Baqlava

Microbial community profiling BAQLaVa
1

Hi,

I have been trying to install baqlava. First I installed humann v4.0.0.alpha.1. However, I am unable to install anadama2 (as a pre-requiry) properly –> doesn’t give any error but doesn’t show any output when hitting anadama2 –help. Do I need to install any specific version of anadama?

Thanks!

Faezeh

2

Hi Faezeh! BAQLaVa is not yet compatible with HUMAnN 4, please install v3.9. It is in the works to get a 4.0 compatible version so we should have that soon but just not yet! Is your AnADAMA2>=0.10.0?

3

Hi Jensen,

Thanks a lot for your prompt response! Yes, I tried different versions of AnADAMA2 and all were the same.

Regarding HUMAnN v3.9, would you please guide me with all the database that is compatible with it? I struggled a lot for that too and at the end found the solution in this topic:

4

Hi @Faeze_Darbaniyan ,

Thank you so much for reaching out to the bioBakery Lab. Re: For the anadama2 issue, I do not think the command anadama2 –-help will show the installed version or help options.

Can you type the following commands to test if anadama2 is installed in your system?

python
import anadama2
print(anadama2.version)

Regards,
Sagun

5

Hi Sagun,

Thanks a lot for your prompt response!

When I try to install it using $ conda install -c bioconda anadama2 inside the environment that I have installed HUMAaN 4, everything looks like to be working: # All requested packages already installed.

However, when I try $import anadama2 I get ModuleNotFoundError: No module named ‘leveldb’

And for print(anadama2.version) I get NameError: name ‘anadama2’ is not defined

Would you please guide me from scratch if I need to do it in a different conda environment. The question is which database is compatible with HUMAaN older versions if I have to use older HUMAaN version.

Thanks!

Faezeh

6

Dear @sagunmaharjann ,

I just wanted to update that I could overcome the leveldb error by running:

$ conda install -c conda-forge python-leveldb

Then, installed anadama2 using:

$ conda install -c bioconda anadama2

Now, trying the command you suggested gives me:

$ python

Python 3.12.12 | packaged by conda-forge | (main, Oct 22 2025, 23:25:55) [GCC 14.3.0] on linux

Type “help”, “copyright”, “credits” or “license” for more information.

>>> import anadama2

>>> print(anadama2.version)

Traceback (most recent call last):

File “”, line 1, in

AttributeError: module ‘anadama2’ has no attribute ‘version’

Is there something I am missing?

Thanks a lot!

Faezeh

7

Hi @jjensen44 ,

I am still struggling with having this working. I have also tried installing humann 3.9 but faced difficulty with its compatibility with metaphlan… Went through all related topics in the forum but no success. I was wondering if I should expect any updates from you.

Thanks a lot!

Faezeh

8

Hi @Faeze_Darbaniyan ,

Apologies you are still running into the installing issues for baqlava and its dependencies (humann and metaphlan). I see that you are now using humann v3.9 and metaphlan v4.1 with mpa_vJun23_CHOCOPhlAnSGB_202307 database in this post which is the correct path forward for baqlava installation.

Since metaphlan v4.1 demo is running as expected and humann v3.9 demo is not, this error usually happens when HUMAnN can’t properly run metaphlan --version, often because of an environment mismatch or small formatting issues in the command.

Could you quickly run:

which humann
which metaphlan
metaphlan --version

and confirm they’re coming from the same conda environment? Also, please try retyping the HUMAnN command manually just to avoid any hidden special characters(quotes).

Let me know what you see and we’ll sort it out.

Regards,
Sagun

9

Thank you so much @sagunmaharjann for your help!

Here is the output of what you asked for:

(humann3) [fdarbaniyan@ldragon3 ~]$ which humann

~/Rafael/install/ls/envs/humann3/bin/humann

(humann3) [fdarbaniyan@ldragon3 ~]$ which metaphlan

~/Rafael/install/ls/envs/humann3/bin/metaphlan

(humann3) [fdarbaniyan@ldragon3 ~]$ metaphlan --version

MetaPhlAn version 4.1.2 (16 Oct 2025)

No complete MetaPhlAn Bowtie2 database found

And I also retyped everything manually and still receive same error: CRITICAL ERROR: Can not call software version for metaphlan

Thanks!

Faezeh

10

Dear @sagunmaharjann ,

I also wanted to add that my metaphlan index is as follows:

total 23407904
-rw-rw-r-- 1 4389008193 Aug  3  2023 mpa_vJun23_CHOCOPhlAnSGB_202307.1.bt2l
-rw-rw-r-- 1 5634079508 Aug  3  2023 mpa_vJun23_CHOCOPhlAnSGB_202307.2.bt2l
-rw-rw-r-- 1  126529023 Aug  3  2023 mpa_vJun23_CHOCOPhlAnSGB_202307.3.bt2l
-rw-rw-r-- 1 2817039751 Aug  3  2023 mpa_vJun23_CHOCOPhlAnSGB_202307.4.bt2l
-rw-rw-r-- 1   98290559 Nov 27  2023 mpa_vJun23_CHOCOPhlAnSGB_202307.pkl
-rw-rw-r-- 1 4389008193 Aug  3  2023 mpa_vJun23_CHOCOPhlAnSGB_202307.rev.1.bt2l
-rw-rw-r-- 1 5634079508 Aug  3  2023 mpa_vJun23_CHOCOPhlAnSGB_202307.rev.2.bt2l
-rw-rw-r-- 1      44092 Nov 27  2023 mpa_vJun23_CHOCOPhlAnSGB_202307_VINFO.csv
-rwx------ 1  881484571 Feb 20 16:27 mpa_vJun23_CHOCOPhlAnSGB_202307_VSG.fna

Please let me know if I am missing anything.

Thanks!

Faezeh

11

Hello @Faeze_Darbaniyan ,

Thank you for trying the commands and follow up. Everything looks good with the software installation. Can you try with moving the metaphlan database to default path where metaphlan is expecting the databases please?

Example Demo command without having to pass bowtie2 index flag:

$ metaphlan example.fastq --input_type fastq -o profiled_metagenome.txt

Regards,
Sagun

12

Thanks @sagunmaharjann for your reply!

I just moved the metaphlan database to the default path (substituted vJun23 with vJan25 version). Now I have the following as expected:

$ metaphlan --version
MetaPhlAn version 4.1.2 (16 Oct 2025)
Installed databases: mpa_vJun23_CHOCOPhlAnSGB_202307

Trying your command gives me the following error:

ERROR: Cannot find a local database. Please run MetaPhlAn using option "-x <database_name>".
            You can download the MetaPhlAn database from 
 http://cmprod1.cibio.unitn.it/biobakery4/metaphlan_databases

Checking it with database path I get:

$ metaphlan --input_type fastq examples/demo.fastq.gz --bowtie2db humann3_db/metaphlan_db_vJun23 --index mpa_vJun23_CHOCOPhlAnSGB_202307 -o profiled_metagenome.txt
[WARNING] Failed to launch x86-64-v3 version, staying with default
[WARNING] Failed to launch x86-64-v3 version, staying with default
[WARNING] Failed to launch x86-64-v3 version, staying with default
[WARNING] Failed to launch x86-64-v3 version, staying with default
WARNING: The metagenome profile contains clades that represent multiple species merged into a single representant.
An additional column listing the merged species is added to the MetaPhlAn output.

And checking the humann example, I have:

$ humann -i examples/demo.fastq.gz -o examples/SAMPLE2 --threads 16 --protein-database humann3_db/uniref --nucleotide-database humann3_db/chocophlan --metaphlan-options "–bowtie2db humann3_db/metaphlan_db_vJun23 -x mpa_vJun23_CHOCOPhlAnSGB_202307"
Output files will be written to: /rsrch6/home/hema_bio-Malignan/fdarbaniyan/Neeraj/HUMANN/examples/SAMPLE2
Decompressing gzipped file ...

CRITICAL ERROR: Can not call software version for metaphlan

Apparently, metaphlan and humann are still not connected.

Thanks!

Faezeh