hmmpfam component and hmmconvert problems

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

hmmpfam component and hmmconvert problems

ML mail
Dear Ergatis users,

I am still trying to run the prokaryotic annotation pipeline and encountered an
issue with the hmmpfam component.


First of all I have followed Aaron's answer on how to generate the HMM libraries
and described here:

http://gmod.827538.n3.nabble.com/hmmpfam-pre-overlap-analysis-prok-pipeline-tt1534273.html#a1561266%23none



but I am stuck at the very last part, which is to convert the library to a
binary file using hmmconvert. The problem is that I am using hmmer version 2.3.2
and when I run the following command:

$ hmmconvert -b coding_hmm.lib coding_hmm.lib.bin

I get the following error:

FATAL: Failed to open HMM database coding_hmm.lib

but my file coding_hmm.lib really exists, so I suppose hmmconvert from hmmer
version 2.3.2 is not able to cope with more recent HMM library files (actually I
am using Pfam-A R25 and TIGRFAMs R10.1).


So what I did is to download the latest version of hmmer (v3.0) and use the
hmmconvert tool of that version to create the binary library file. Success, this
worked. But now I have another problem: hmmpfam can't use this library because
it's too recent, as you can see in the output here:

WARNING: /db/hmm/coding_hmm.lib.bin appears to be a binary but format is not
recognized
It may be from a HMMER version more recent than yours,
or may be a different kind of binary altogether.

I wanted to use hmmpfam from hmmer v3.0 but unfortunately hmmpfam doesn't exist
anymore in hmmer v3.0.


So to conclude, I am guessing that the hmmpfam component of Ergatis is outdated
and would need an update to use hmmer v3.0, am I correct here?


Or does anyone know a workaround so that I can still use the hmmpfam component?
I really would like to make use of this component.

Many thanks in advance for your feedback.

Regards,
ML


------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ergatis-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ergatis-users
Reply | Threaded
Open this post in threaded view
|

Re: hmmpfam component and hmmconvert problems

ML mail
Just to let you know that I have saw that there is a hmmpfam3 component and
replaced the older hmmpfam component with hmmpfam3 in the prokaryotic annotation
pipeline. Unfortunately, I had to restart my pipeline due to that, so I will let
you know how it goes as soon as it reaches that stage.

Btw: important for people doing the same, you will need to run hmmpress to
create a binary file out of your hmm library (this is the equivalent of
hmmconvert in hmmer v2).






----- Original Message ----
From: ML mail <[hidden email]>
To: [hidden email]
Sent: Wed, May 11, 2011 11:10:13 AM
Subject: [Ergatis-users] hmmpfam component and hmmconvert problems

Dear Ergatis users,

I am still trying to run the prokaryotic annotation pipeline and encountered an
issue with the hmmpfam component.


First of all I have followed Aaron's answer on how to generate the HMM libraries

and described here:

http://gmod.827538.n3.nabble.com/hmmpfam-pre-overlap-analysis-prok-pipeline-tt1534273.html#a1561266%23none




but I am stuck at the very last part, which is to convert the library to a
binary file using hmmconvert. The problem is that I am using hmmer version 2.3.2

and when I run the following command:

$ hmmconvert -b coding_hmm.lib coding_hmm.lib.bin

I get the following error:

FATAL: Failed to open HMM database coding_hmm.lib

but my file coding_hmm.lib really exists, so I suppose hmmconvert from hmmer
version 2.3.2 is not able to cope with more recent HMM library files (actually I

am using Pfam-A R25 and TIGRFAMs R10.1).


So what I did is to download the latest version of hmmer (v3.0) and use the
hmmconvert tool of that version to create the binary library file. Success, this

worked. But now I have another problem: hmmpfam can't use this library because
it's too recent, as you can see in the output here:

WARNING: /db/hmm/coding_hmm.lib.bin appears to be a binary but format is not
recognized
It may be from a HMMER version more recent than yours,
or may be a different kind of binary altogether.

I wanted to use hmmpfam from hmmer v3.0 but unfortunately hmmpfam doesn't exist
anymore in hmmer v3.0.


So to conclude, I am guessing that the hmmpfam component of Ergatis is outdated
and would need an update to use hmmer v3.0, am I correct here?


Or does anyone know a workaround so that I can still use the hmmpfam component?
I really would like to make use of this component.

Many thanks in advance for your feedback.

Regards,
ML


------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ergatis-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ergatis-users


------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ergatis-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ergatis-users
Reply | Threaded
Open this post in threaded view
|

Re: hmmpfam component and hmmconvert problems

gary_xie
Hi, ML and all,
I thought hmmpress is to Format an HMM database into a binary format  
for hmmscan (based on hmmer user guide), but not for hmmpfam. The  
hmmpfam executable is not distributed as a part of the hmmer3  
package,  but we can download and extract from iprscan package, I am  
assuming this is the older version v2.
ftp://ftp.ebi.ac.uk/pub/software/unix/iprscan/BIN/4.x/
I have noticed hmmpfam and hmmconvert, hmmpress and hmmscan are all  
included in iprscab package.
My question is : can we still use hmmpfam  executable (from iprscan)  
with Ergatis coonfig components (xml and config file) to search  
against hmmconvert -b builded binary hmmer format or hmmpress builded  
binary format. Which version of hmmpfam Ergatis component I should  
config in software. config? the hmmpfam or hmmpfam3? Both xml and  
config files are available.
Anyone has  insight or willing to help?
Thanks,
Gary

On May 11, 2011, at 5:45 AM, ML mail wrote:

> Just to let you know that I have saw that there is a hmmpfam3  
> component and
> replaced the older hmmpfam component with hmmpfam3 in the  
> prokaryotic annotation
> pipeline. Unfortunately, I had to restart my pipeline due to that,  
> so I will let
> you know how it goes as soon as it reaches that stage.
>
> Btw: important for people doing the same, you will need to run  
> hmmpress to
> create a binary file out of your hmm library (this is the equivalent  
> of
> hmmconvert in hmmer v2).
>
>
>
>
>
>
> ----- Original Message ----
> From: ML mail <[hidden email]>
> To: [hidden email]
> Sent: Wed, May 11, 2011 11:10:13 AM
> Subject: [Ergatis-users] hmmpfam component and hmmconvert problems
>
> Dear Ergatis users,
>
> I am still trying to run the prokaryotic annotation pipeline and  
> encountered an
> issue with the hmmpfam component.
>
>
> First of all I have followed Aaron's answer on how to generate the  
> HMM libraries
>
> and described here:
>
> http://gmod.827538.n3.nabble.com/hmmpfam-pre-overlap-analysis-prok-pipeline-tt1534273.html#a1561266%23none
>
>
>
>
> but I am stuck at the very last part, which is to convert the  
> library to a
> binary file using hmmconvert. The problem is that I am using hmmer  
> version 2.3.2
>
> and when I run the following command:
>
> $ hmmconvert -b coding_hmm.lib coding_hmm.lib.bin
>
> I get the following error:
>
> FATAL: Failed to open HMM database coding_hmm.lib
>
> but my file coding_hmm.lib really exists, so I suppose hmmconvert  
> from hmmer
> version 2.3.2 is not able to cope with more recent HMM library files  
> (actually I
>
> am using Pfam-A R25 and TIGRFAMs R10.1).
>
>
> So what I did is to download the latest version of hmmer (v3.0) and  
> use the
> hmmconvert tool of that version to create the binary library file.  
> Success, this
>
> worked. But now I have another problem: hmmpfam can't use this  
> library because
> it's too recent, as you can see in the output here:
>
> WARNING: /db/hmm/coding_hmm.lib.bin appears to be a binary but  
> format is not
> recognized
> It may be from a HMMER version more recent than yours,
> or may be a different kind of binary altogether.
>
> I wanted to use hmmpfam from hmmer v3.0 but unfortunately hmmpfam  
> doesn't exist
> anymore in hmmer v3.0.
>
>
> So to conclude, I am guessing that the hmmpfam component of Ergatis  
> is outdated
> and would need an update to use hmmer v3.0, am I correct here?
>
>
> Or does anyone know a workaround so that I can still use the hmmpfam  
> component?
> I really would like to make use of this component.
>
> Many thanks in advance for your feedback.
>
> Regards,
> ML
>
>
> ------------------------------------------------------------------------------
> Achieve unprecedented app performance and reliability
> What every C/C++ and Fortran developer should know.
> Learn how Intel has extended the reach of its next-generation tools
> to help boost performance applications - inlcuding clusters.
> http://p.sf.net/sfu/intel-dev2devmay
> _______________________________________________
> Ergatis-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/ergatis-users
>
>
> ------------------------------------------------------------------------------
> Achieve unprecedented app performance and reliability
> What every C/C++ and Fortran developer should know.
> Learn how Intel has extended the reach of its next-generation tools
> to help boost performance applications - inlcuding clusters.
> http://p.sf.net/sfu/intel-dev2devmay
> _______________________________________________
> Ergatis-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/ergatis-users


------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ergatis-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ergatis-users
Reply | Threaded
Open this post in threaded view
|

Re: hmmpfam component and hmmconvert problems

ML mail
Hi Gary and Ergatis fellows,

You are totally right, hmmpress is to format hmm v3 databases into binaries
which then can only be read by hmmer v3. hmmpfam has been removed in the newer
version and it's equivalent is hmmscan.

The problem here is that you can't use anymore hmmer v2 with recent databases
like Pfam-A or TIGRFAMs because they all use HMM v3 formatting and hmmconvert
won't recognized that new format. So if you are going to use an old already
formatted HMM library it's fine to use the hmmpfam component but if you are like
me with a fresh new installation of Ergatis and up-to-date downloaded Pfam-A and
TIGRFAMs then you will need to replace the hmmpfam component with hmmpfam3.

I would suggest in the future version of Ergatis to generally replace hmmpfam
with hmmpfam3 component in all the pipelines using it (I am currently using the
prokaryotic annotation pipeline). I have just tested it with this replacement
and it worked fine but now I am stuck at blastx which takes ages (don't have a
cluster here so I have to be patient...). My only concern is if the output of
the hmmpfam3 component will be compatible with the latest steps of that very
same pipeline as the output formatting might have changed, but I will see that
when I arrive at this stage.

Regards,
ML





----- Original Message ----
From: Gary Xie <[hidden email]>
To: [hidden email]
Sent: Thu, May 12, 2011 9:39:39 PM
Subject: Re: [Ergatis-users] hmmpfam component and hmmconvert problems

Hi, ML and all,
I thought hmmpress is to Format an HMM database into a binary format  
for hmmscan (based on hmmer user guide), but not for hmmpfam. The  
hmmpfam executable is not distributed as a part of the hmmer3  
package,  but we can download and extract from iprscan package, I am  
assuming this is the older version v2.
ftp://ftp.ebi.ac.uk/pub/software/unix/iprscan/BIN/4.x/
I have noticed hmmpfam and hmmconvert, hmmpress and hmmscan are all  
included in iprscab package.
My question is : can we still use hmmpfam  executable (from iprscan)  
with Ergatis coonfig components (xml and config file) to search  
against hmmconvert -b builded binary hmmer format or hmmpress builded  
binary format. Which version of hmmpfam Ergatis component I should  
config in software. config? the hmmpfam or hmmpfam3? Both xml and  
config files are available.
Anyone has  insight or willing to help?
Thanks,
Gary

On May 11, 2011, at 5:45 AM, ML mail wrote:

> Just to let you know that I have saw that there is a hmmpfam3  
> component and
> replaced the older hmmpfam component with hmmpfam3 in the  
> prokaryotic annotation
> pipeline. Unfortunately, I had to restart my pipeline due to that,  
> so I will let
> you know how it goes as soon as it reaches that stage.
>
> Btw: important for people doing the same, you will need to run  
> hmmpress to
> create a binary file out of your hmm library (this is the equivalent  
> of
> hmmconvert in hmmer v2).
>
>
>
>
>
>
> ----- Original Message ----
> From: ML mail <[hidden email]>
> To: [hidden email]
> Sent: Wed, May 11, 2011 11:10:13 AM
> Subject: [Ergatis-users] hmmpfam component and hmmconvert problems
>
> Dear Ergatis users,
>
> I am still trying to run the prokaryotic annotation pipeline and  
> encountered an
> issue with the hmmpfam component.
>
>
> First of all I have followed Aaron's answer on how to generate the  
> HMM libraries
>
> and described here:
>
>http://gmod.827538.n3.nabble.com/hmmpfam-pre-overlap-analysis-prok-pipeline-tt1534273.html#a1561266%23none
>e
>
>
>
>
> but I am stuck at the very last part, which is to convert the  
> library to a
> binary file using hmmconvert. The problem is that I am using hmmer  
> version 2.3.2
>
> and when I run the following command:
>
> $ hmmconvert -b coding_hmm.lib coding_hmm.lib.bin
>
> I get the following error:
>
> FATAL: Failed to open HMM database coding_hmm.lib
>
> but my file coding_hmm.lib really exists, so I suppose hmmconvert  
> from hmmer
> version 2.3.2 is not able to cope with more recent HMM library files  
> (actually I
>
> am using Pfam-A R25 and TIGRFAMs R10.1).
>
>
> So what I did is to download the latest version of hmmer (v3.0) and  
> use the
> hmmconvert tool of that version to create the binary library file.  
> Success, this
>
> worked. But now I have another problem: hmmpfam can't use this  
> library because
> it's too recent, as you can see in the output here:
>
> WARNING: /db/hmm/coding_hmm.lib.bin appears to be a binary but  
> format is not
> recognized
> It may be from a HMMER version more recent than yours,
> or may be a different kind of binary altogether.
>
> I wanted to use hmmpfam from hmmer v3.0 but unfortunately hmmpfam  
> doesn't exist
> anymore in hmmer v3.0.
>
>
> So to conclude, I am guessing that the hmmpfam component of Ergatis  
> is outdated
> and would need an update to use hmmer v3.0, am I correct here?
>
>
> Or does anyone know a workaround so that I can still use the hmmpfam  
> component?
> I really would like to make use of this component.
>
> Many thanks in advance for your feedback.
>
> Regards,
> ML
>
>
> ------------------------------------------------------------------------------
> Achieve unprecedented app performance and reliability
> What every C/C++ and Fortran developer should know.
> Learn how Intel has extended the reach of its next-generation tools
> to help boost performance applications - inlcuding clusters.
> http://p.sf.net/sfu/intel-dev2devmay
> _______________________________________________
> Ergatis-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/ergatis-users
>
>
> ------------------------------------------------------------------------------
> Achieve unprecedented app performance and reliability
> What every C/C++ and Fortran developer should know.
> Learn how Intel has extended the reach of its next-generation tools
> to help boost performance applications - inlcuding clusters.
> http://p.sf.net/sfu/intel-dev2devmay
> _______________________________________________
> Ergatis-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/ergatis-users


------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ergatis-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ergatis-users


------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ergatis-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ergatis-users
Reply | Threaded
Open this post in threaded view
|

Re: questions about Ergatis components installation

Aaron Buechlein
In reply to this post by gary_xie
Hi Gary,
I think I can help with your third question.  Here's what I do for us at
the CGB.

    First I download TIGRFAM and Pfam-A using the urls below.
    ftp://ftp.jcvi.org/pub/data/TIGRFAMs/TIGRFAMs_10.0_HMM.LIB.gz
    ftp://ftp.sanger.ac.uk/pub/databases/Pfam/current_release/Pfam-A.hmm.gz

    Then I decompress them and cat them together into coding_hmm.lib
    >gunzip TIGRFAMs_10.0_HMM.LIB.gz
    >gunzip Pfam\-A.hmm.gz
    >cat Pfam\-A.hmm TIGRFAMs_10.0_HMM.LIB > coding_hmm.lib

    The last step is creating the .bin file for hmmpfam
    >hmmconvert -b coding_hmm.lib coding_hmm.lib.bin

    Also, here is some additional information regarding the HMM library
    I got in an email for Joshua a while back.  Quoting Joshua:
    "If you want to use the same HMM library set up we do you should
    have the tools to generate it.  First, we do combine the Pfam and
    TIGRFAM collections into a single library.  Then, we use a few
    scripts in the 'ergatis/util' SVN directory to create a MLDBM data
    structure based off it.  The idea is that when BSML is created from
    HMM search results this lookup db provides annotation assertions (GO
    terms, EC numbers, etc.) based on the matched HMM accession.  The
    first script creates the db (hmmlib_to_mldbm.pl) and other scripts
    can add supplemental data to it ( *_to_mldbm.pl ).  Each has perldoc
    to get a sense of what they do."


Regarding the issues with tigr_roles, I do not think those are used for
anything in the current prokaryotic pipeline although someone else may
want to confirm this. So, at the CGB, we just comment them out in the
bin/post_process_annotation.pl file.

    line 69 #use PFunc::TIGRRolesAnnotation;
    line 112     #my ($tigr_roles, $tr_source, $tr_source_type) =
    $annotation->get_TIGR_Role;        
    line 113     #my
    $new_tigr_roles;                                                                

    line 114     #if( $new_tigr_roles =
    $tigr_roles_annot->assign_tigr_roles_by_keyword( $gene_product
    line 115     #    $annotation->set_TIGR_Role(
    $tigr_roles_annot->assign_tigr_roles_by_keyword( $ge
    line 116     #                                'by_keyword',
    'by_keyword' );                      
    line 117     #}    

    line 183     #if( $options{'tigr_roles_db_dir'} ) {
    line 184     #    $tigr_roles_annot = new
    PFunc::TIGRRolesAnnotation( 'tigr_roles_db_dir' =>
    $options{'tigr_roles_db_dir'} );
    line 185     #} else {
    line 186     #    die("Option --tigr_roles_db_dir is required");
    line 187     #}

Hope this helps.
Thanks,

Aaron

Gary Xie wrote:

> Date: Wed, 11 May 2011 16:23:19 -0600
> From: Gary Xie <[hidden email]>
> Reply-To: [hidden email]
> To: [hidden email]
> Subject: [Ergatis-users] questions about Ergatis components installation
>
> Hi, there,
> I am new on Ergatis, so please bear with some my naive questions. We
> are currently doing Ergatis components installation and encounter
> following roadblocks.
>
> 1) In order to install and config p_func component, the tigr_roles db
> needs to be installed and modify path in software.config file.
> [component p_func]
> $;TIGR_ROLES_DB_DIR$;=/local/db/by_source/jcvi/tigr_roles
>
> Since the download link for tigr_roles db was not documented,  I did
> the google search, the closest one i can find is the tigrfam2role.db
> (unix executable file) inside of the lookup table fold of Manatee
> download site. Is this the right file?
> http://sourceforge.net/projects/manatee/files/igs_manatee/2.22.2/
>
> 2) In order to install and config parse_evidence component ,  TIGRFAMs
> needs to be installed and downloaded from
> ftp://ftp.jcvi.org/pub/data/TIGRFAMs/
> [component parse_evidence]
> $;TIGR_ROLES_DB_DIR$;=/local/db/by_source/jcvi/tigr_roles
> $;TIGRFAMS_DIR$;=/local/db/by_source/jcvi/TIGRFAMs/latest
>
> There are a lot of links for version 10, do I need to get them all?
>   is it ok to leave the databases compressed (.gz), or should I unzip
> them?
> For example:  Can the tar files containing SEED and HMM files stay tar
> files? Or do they all have to be untarred?  And if untarred, all to
> the same directory? or is there a directory structure.  I couldn't
> find any good documentation.
>
> 3)Here in a paragraph in software.config file
>   ;; This is a collection of HMMs.  "coding_hmm" is a concatenation of
> ;;  PFams, TIGRFams and select others.
> $;HMM_ALL$;=/mnt/staging/db/coding_hmm/coding_hmm.lib.bin
> $;HMM_LIB_DB$;=/mnt/staging/db/coding_hmm/coding_hmm.lib.db
>
> Can someone explain where to download coding_hmm.lib.bin and
> coding_hmm/coding_hmm.lib.db and any instruction on how to install them?
>
> 4) j_ortholog_clusters componment is needed for pangenome pipeline.
> Where can i download it?
>
> Can someone help me out or point out the right direction...
> Thanks,
> Gary
>  
>
>
> ------------------------------------------------------------------------------
> Achieve unprecedented app performance and reliability
> What every C/C++ and Fortran developer should know.
> Learn how Intel has extended the reach of its next-generation tools
> to help boost performance applications - inlcuding clusters.
> http://p.sf.net/sfu/intel-dev2devmay
> _______________________________________________
> Ergatis-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/ergatis-users
>  

------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ergatis-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ergatis-users
Reply | Threaded
Open this post in threaded view
|

Ergatis sub-templates not loading properly

Chris Hemmerich
In reply to this post by ML mail

Hi,

  I'm testing the new pipelines in r14 and while templates load components
normally, they don't load other templates. I never tried this feature
under r13, so my problem may not be new.

Checking the apache logs, for a component-only template, I see a hit to
build_pipeline.cgi, then get_pipeline_layout.cgi, then calls to
get_component_template.cgi and save_component.cgi for each component and
finally run_pipeline.cgi.

For a template-only template, I get build_pipeline.cgi,
get_pipeline_layout.cgi, and then straight to run_pipeline.cgi which fails
because the the parent directory for bulding the pipeline hasn't been
created.

I made sure I'm using web files from the new release, and since it looks
like the calls are coming from the browser, I made sure I new local copies
of all the js files and tried Safari and Chrome, but all had the same
issue.

Any thoughts on where I've gone wrong, or where I should look next?

Thanks,
  Chris



------------------------------------------------------------------------------
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
_______________________________________________
Ergatis-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/ergatis-users