From yangpc at biols.ac.cn  Sun Aug 19 12:21:13 2018
From: yangpc at biols.ac.cn (Pengcheng Yang)
Date: Sun, 19 Aug 2018 18:21:13 +0800
Subject: [Inparanoid] local run inparanoid_4.1 found only 5779 mouse protein
 are orthologs of human
Message-ID: <de4c52bf-22d3-1597-574b-78fcfc155ec6@biols.ac.cn>

Dear colleagues,

I have run Inparanoid_4.1 between my studied organism and human. As 
comparison, I run the same program between mouse and human. These two 
runs used the same human protein data set and same program, but with 
different input species A. But the identified ortholog in mouse was only 
5779, it is very strange. So I run the same program with Rat and human 
as input species A and B. The Rat displayed 16979 orthologs of human, 
which is reasonable.

1. I have checked the blast results and found there are  21745 
(mouse.pep-human.fa), 21733 (human.fa-mouse.pep), and 18017 
(mouse.pep-mouse.pep) unique mouse protein IDs. These results suggested 
that the blast step is ok.

2. I checked the message output from inparanoid.pl, and found the 
warning message from mouse are 27849 lines, in contrast, only 439 lines 
from rat. It seems this problem was originated from the "ortholog 
detection" step.

I have rerun the program several times, but with the same results. 
Because I used the exactly same program with same parameters, I don't 
know how to debug this problem. Could anyone help me out?

Best,

Pengcheng Yang


From erik.sonnhammer at scilifelab.se  Thu Aug 30 13:39:08 2018
From: erik.sonnhammer at scilifelab.se (Erik Sonnhammer)
Date: Thu, 30 Aug 2018 13:39:08 +0200
Subject: [Inparanoid] local run inparanoid_4.1 found only 5779 mouse
 protein are orthologs of human
In-Reply-To: <de4c52bf-22d3-1597-574b-78fcfc155ec6@biols.ac.cn>
References: <de4c52bf-22d3-1597-574b-78fcfc155ec6@biols.ac.cn>
Message-ID: <c2867631-92cb-9f7c-9e20-32a27d621eb6@scilifelab.se>

Hi Pengcheng,

what are those warning messages?  What was the source of the mouse 
proteome?  My guess is that something is wrong with it. Please check 
what happens with the mouse proteome from 
http://inparanoid.sbc.su.se/download/8.0_current/sequences/processed/.

/Erik

On 2018-08-19 12:21, Pengcheng Yang wrote:
> Dear colleagues,
>
> I have run Inparanoid_4.1 between my studied organism and human. As 
> comparison, I run the same program between mouse and human. These two 
> runs used the same human protein data set and same program, but with 
> different input species A. But the identified ortholog in mouse was 
> only 5779, it is very strange. So I run the same program with Rat and 
> human as input species A and B. The Rat displayed 16979 orthologs of 
> human, which is reasonable.
>
> 1. I have checked the blast results and found there are  21745 
> (mouse.pep-human.fa), 21733 (human.fa-mouse.pep), and 18017 
> (mouse.pep-mouse.pep) unique mouse protein IDs. These results 
> suggested that the blast step is ok.
>
> 2. I checked the message output from inparanoid.pl, and found the 
> warning message from mouse are 27849 lines, in contrast, only 439 
> lines from rat. It seems this problem was originated from the 
> "ortholog detection" step.
>
> I have rerun the program several times, but with the same results. 
> Because I used the exactly same program with same parameters, I don't 
> know how to debug this problem. Could anyone help me out?
>
> Best,
>
> Pengcheng Yang
>
> _______________________________________________
> InParanoid-at-sbc.su.se mailing list
> InParanoid-at-sbc.su.se at lists.su.se
> https://lists.su.se/mailman/listinfo/inparanoid-at-sbc.su.se