Re: [MITObim-users] miraconvert seems to have failed

[chrisi.hahni]

Hi Regina,

It looks like you're using mira 4.9.5. mitobim 1.8 is compatible with mira 4.0.x.

Hope that helps!

Cheers,

Christoph 

-------- Original message --------

From: regina.lo...@gmail.com

Date: 04/12/2016 10:09 (GMT+01:00)

To: MITObim-users <mitobi...@googlegroups.com>

Subject: [MITObim-users] miraconvert seems to have failed

Hi Christoph,

I am trying to reconstruct mitochondrial genomes from IonTorrent NGS data and as I have a reference genome from to the same species I thought of using the --quick approach (output message below). Apparently, my readpool does not contain any reads with reasonable match (kmer=31), which is quite surprising considering that I’m using a reference genome of the same species.
Then, I decided to run the two-step approach and after successfully mapping assembly with MIRA 4, I tried to run MITObim and got the below error message saying that miraconvert seems to have failed - see detailed output below. Any idea how to move forward?

Many thanks,
Regina L. Cunha

--quick option:
MITObim - mitochondrial baiting and iterative mapping
version 1.8
author: Christoph Hahn, (c) 2012-2016

quick option selected! -maf option will be ignored (if given)

./MITObim_1.8.pl -start 1 -end 30 -sample testpool -ref perna_mt_genome -readpool /home/regina/mtDNA/BZP.fastq --quick /home/regina/mtDNA/pernamtDNA.fa

All paramters seem to make sense:
startiteration: 0
enditeration: 30
sample: testpool
refname: perna_mt_genome
readpool: /home/regina/mtDNA/BZP.fastq
maf: 0
quick: /home/regina/mtDNA/pernamtDNA.fa
paired: 0 (off=0, on=1)
assembly mode: 0 (mapping=0, denovo=1)
verbose: 0 (off=0, on=1)
split: 0 (off=0, on=1)
minimum avg. coverage: 0 (off=0)
clean: 0 (off=0, on=1)
trim reads: 0 (off=0, on=1)
trim overhang: 0 (no=0, yes=1)
platform: iontorrent
kmer baiting: 31

Starting MITObim

==============
ITERATION 0

Dec 3 13:03:32

quick option baits reads from provided reference in iteration 0

fishing readpool using mirabait (k = 31)

mirabait: No bait files defined via -b and no -L given!
Did you use the command line for the old mirabait (<= 4.0.2)?

your readpool does not contain any reads with reasonable match (k = 31) to your reference - Maybe you want to different settings or even a different reference?

2-step option:
./MITObim_1.8.pl -start 1 -end 30 -sample testpool -ref perna_mt_genome -readpool /home/regina/mtDNA/BZP.fastq -maf /home/regina/mtDNA/initial-mapping-testpool-to-perna-mt_assembly/initial-mapping-testpool-to-perna-mt_d_results/initial-mapping-testpool-to-perna-mt_out.maf

All paramters seem to make sense:
startiteration: 1
enditeration: 30
sample: testpool
refname: perna_mt_genome
readpool: /home/regina/mtDNA/BZP.fastq
maf: /home/regina/mtDNA/initial-mapping-testpool-to-perna-mt_assembly/initial-mapping-testpool-to-perna-mt_d_results/initial-mapping-testpool-to-perna-mt_out.maf
quick: 0
paired: 0 (off=0, on=1)
assembly mode: 0 (mapping=0, denovo=1)
verbose: 0 (off=0, on=1)
split: 0 (off=0, on=1)
minimum avg. coverage: 0 (off=0)
clean: 0 (off=0, on=1)
trim reads: 0 (off=0, on=1)
trim overhang: 0 (no=0, yes=1)
platform: iontorrent
kmer baiting: 31

Starting MITObim

==============
ITERATION 1

Dec 3 13:19:02

recover backbone by running miraconvert on maf file

ConvPro::checkTypes: Unknown or illegal file type 'tmp' defined as
miraconvert (MIRALIB version 4.9.5_2 )
Author: Bastien Chevreux (ba...@chevreux.org)
Purpose: convert assembly and sequencing file types.

Usage: [......]
miraconvert seems to have failed - see detailed output above

--
You received this message because you are subscribed to the Google Groups "MITObim-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mitobim-user...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[regina.lo...@gmail.com]

Hi Christoph,

I followed your suggestion and re-installed mira with the version 4.0.2 but now I am getting the following message in MITObim, which is the same I got before with the --quick option. This is quite strange because the reference is a mtDNA genome of the same species so couldn't be closer! And the quality report from the raw reads was ok (The main goal of this project is to sequence the mtDNA of 5 samples with a mean depth coverage of 80-100x, assuming a mitochondrial genome size of 18,4 kbp. The produced sequence data is in accordance with the expected output since more than 45 Mpb, mean depth coverage of ≥ 2 445 x, was generated per each sample).

Any advice would be really appreciated
Regina

Full command run:
./MITObim_1.8.pl --mirapath /home/regina/mtDNA/MITObim/mira_4.0.2 -start 1 -end 30 -sample testpool -ref perna_mt_genome -readpool /home/regina/mtDNA/PDO.fastq -maf /home/regina/mtDNA/MITObim/mira_4.0.2/initial-mapping-testpool-to-perna-mt_assembly/initial-mapping-testpool-to-perna-mt_d_results/initial-mapping-testpool-to-perna-mt_out.maf

All paramters seem to make sense:

All paramters seem to make sense:
startiteration: 1
enditeration: 30
sample: testpool
refname: perna_mt_genome

readpool: /home/regina/mtDNA/PDO.fastq
maf: /home/regina/mtDNA/MITObim/mira_4.0.2/initial-mapping-testpool-to-perna-mt_assembly/initial-mapping-testpool-to-perna-mt_d_results/initial-mapping-testpool-to-perna-mt_out.maf

quick: 0
paired: 0 (off=0, on=1)
assembly mode: 0 (mapping=0, denovo=1)
verbose: 0 (off=0, on=1)
split: 0 (off=0, on=1)
minimum avg. coverage: 0 (off=0)
clean: 0 (off=0, on=1)
trim reads: 0 (off=0, on=1)
trim overhang: 0 (no=0, yes=1)
platform: iontorrent
kmer baiting: 31

Starting MITObim

==============
 ITERATION 1

==============
Dec  5 12:42:50

recover backbone by running miraconvert on maf file

fishing readpool using mirabait (k = 31)

your readpool does not contain any reads with reasonable match (k = 31) to your reference - Maybe you ll want to different settings or even a different reference?

[regina.lo...@gmail.com]

Me again,

just ran the --quick version again with the --verbose option. The temp_baitfile.fasta file from the iteration0 seems to be ok (18,415bp). The output of the initial-mapping-testpool-to-perna-mt_out_testpool.unpadded.fasta is not ok (I copied below a part of it but it is al xxx's till the end).
See also below the output of the run with the --quick and --verbose option.

Many thanks,
Regina


>KM655841.1_bb
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
[....]


Output of the MITObim of the run with the --quick and --verbose option:

regina@ccmar-r2c2-01:~/mtDNA/MITObim> ./MITObim_1.8.pl --mirapath /home/regina/mtDNA/MITObim/mira_4.0.2 --verbose -start 1 -end 30 -sample testpool -ref perna_mt_genome -readpool /home/regina/mtDNA/BZP.fastq --quick /home/regina/mtDNA/pernamtDNA.fa
perl: warning: Setting locale failed.
perl: warning: Please check that your locale settings:
        LANGUAGE = "en_US:en",
        LC_ALL = (unset),
        LC_CTYPE = "UTF-8",
        LANG = "en_US.UTF-8"
    are supported and installed on your system.
perl: warning: Falling back to a fallback locale ("en_US.UTF-8").

MITObim - mitochondrial baiting and iterative mapping
version 1.8
author: Christoph Hahn, (c) 2012-2016

quick option selected! -maf option will be ignored (if given)

found executables in the path specified by the user - good!

Full command run:
./MITObim_1.8.pl --mirapath /home/regina/mtDNA/MITObim/mira_4.0.2 --verbose -start 1 -end 30 -sample testpool -ref perna_mt_genome -readpool /home/regina/mtDNA/BZP.fastq --quick /home/regina/mtDNA/pernamtDNA.fa

All paramters seem to make sense:
startiteration: 0
enditeration: 30
sample: testpool
refname: perna_mt_genome
readpool: /home/regina/mtDNA/BZP.fastq
maf: 0
quick: /home/regina/mtDNA/pernamtDNA.fa
paired: 0 (off=0, on=1)
assembly mode: 0 (mapping=0, denovo=1)

verbose: 1 (off=0, on=1)

split: 0 (off=0, on=1)
minimum avg. coverage: 0 (off=0)
clean: 0 (off=0, on=1)
trim reads: 0 (off=0, on=1)
trim overhang: 0 (no=0, yes=1)
platform: iontorrent
kmer baiting: 31

Starting MITObim

==============
 ITERATION 0

==============
Dec  5 14:45:52

quick option baits reads from provided reference in iteration 0

fishing readpool using mirabait (k = 31)

Your system seems to be older or have some quirks with locale settings.
 Using the LC_ALL=C workaround.
 If you don't want that, fix your system ;-)
 /home/regina/mtDNA/BZP.fastq 1 /home/regina/mtDNA 2 BZP 3 fastq
 Loading baits ...Localtime: Mon Dec  5 14:45:52 2016
 Loading data from FASTA file:
  [0%] ....|.... [10%] ....|.... [20%] ....|.... [30%] ....|.... [40%] ....|.... [50%] ....|.... [60%] ....|.... [70%] ....|.... [80%] ....|.... [90%] ....|.... [100%]
 Localtime: Mon Dec  5 14:45:52 2016
 rnm size: 0
 No FASTA quality file given, using default qualities for all reads just loaded.
 Localtime: Mon Dec  5 14:45:52 2016
 
 Done.
 Loaded 1 reads with 0 reads having quality accounted for.
 baitrp.size(): 1
 Localtime: Mon Dec  5 14:45:52 2016
 Writing temporary hstat files:
 freemem: 3282542592
 TNH: 18385
 XME 1: 0.000730554
 XME 2: 0.1
 NEPB 1: 104857
 NEPB 2: 104857
  [0%] ....|.... [10%] ....|.... [20%] ....|.... [30%] ....|.... [40%] ....|.... [50%] ....|.... [60%] ....|.... [70%] ....|.... [80%] ....|.... [90%] ....|.... [100%] done
 Flushing buffers to disk:
  [0%] ....|.... [10%] ....|.... [20%] ....|.... [30%] ....|.... [40%] ....|.... [50%] ....|.... [60%] ....|.... [70%] ....|.... [80%] ....|.... [90%] ....|.... [100%] done
 Localtime: Mon Dec  5 14:45:52 2016
 
 Analysing hstat files:
  [0%] ....|.... [10%] ....|.... [20%] ....|.... [30%] ....|.... [40%] ....|.... [50%] ....|.... [60%] ....|.... [70%] ....|.... [80%] ....|.... [90%] ....|.... [100%]
 Localtime: Mon Dec  5 14:45:52 2016
 clean up temporary stat files...Localtime: Mon Dec  5 14:45:52 2016
 Raw MHI: 18374
 Raw avg. freq. : 1
 HSS 36748      HSST: 33074
 Localtime: Mon Dec  5 14:45:52 2016
 Will save to: fastq
 Loading data from fastq ...Localtime: Mon Dec  5 14:45:52 2016
 Loading data from FASTQ file: /home/regina/mtDNA/BZP.fastq
 (sorry, no progress indicator for that, possible only with zlib >=1.34)
 
 
 Done.
 Loaded 330614 reads, Localtime: Mon Dec  5 14:45:59 2016
  done.
 
 Baiting process finished.
 
 Number of bait sequences:   1
 Number of sequences baited: 0 / 330614 (0.00%)

[Regina Cunha]

Hi Christoph,

just a quick question: how can we detect differences in gene order in mitogenomes generated by MITObim if we provide a reference?

Cheers,

Regina

On 05 Dec 2016, at 13:21, Christoph Hahn <C.H...@hull.ac.uk> wrote:

Hi Regina,

That's indeed a bit strange. If you are using a mt genome of the same species, you won't need the initial mapping assembly. You might as well just use the `--quick` option. Anyway, now that you've already done it, can you please double-check that a bait file has been created and maybe look into it to check if it contains a reasonable sequence. The file should be called `temp_baitfile.fasta` in your iteration 1 directory. This is the fasta sequence (without ambiguities) extracted from the result (maf file) of the initial assembly. Could be something has gone wrong already there. Maybe the file is empty or otherwise weird, that would explain why no reads seem to be baited. Check out also the result from the initial mapping assembly. Should be in the directory called `your project_d_results` and the file should be called `your_project_out.unpadded.fasta` or something like that.

In parallel can you try to rerun MITObim with mira 4.0.2 and the `--quick` option? I know you already tried, but the this time please specify the verbose (`-v`), which should result in more detailed output from the baiting step. Can you save the output to a txt file and send it over, in case it fails?

Thanks!

Cheers,

Christoph

---

From: regina.lo...@gmail.com [regina.lo...@gmail.com]
Sent: 05 December 2016 12:52
To: MITObim-users
Subject: [MITObim-users] Re: miraconvert seems to have failed

-- 
You received this message because you are subscribed to the Google Groups "MITObim-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mitobim-user...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

**************************************************
To view the terms under which this email is 
distributed, please go to 
http://www2.hull.ac.uk/legal/disclaimer.aspx
**************************************************

[Chris H]

Hi,

MITObim is only built to do assemblies and I am afraid it does not have functionality to look for gene rearrangements. You will have to use a different tool to look into things like that. Once you have a mt genome assembly, the first step would be to predict genes in the assembly. There is a bunch of pipelines out there for that, e.g. MITOS, DOGMA, and then you can take it from there. Changes in gene order/rearrangements should then be pretty obvious. You could try Mauve (http://darlinglab.org/mauve/mauve.html), to detect/visualize rearrangements not necessarily specific to genes, but still. Off the top of my head, I am afraid, I don't recall any tool that specifically looks for gene rearrangements in mt genomes, but I wouldn't be surprised if something exists out there. 

If you find any, I, and possible other users, would be happy to hear about it. Maybe you could give us an update - Thanks!

cheers,
Christoph 

[Chris H]

Hi,

The discussion about the Regina's problems with read mapping continues here.

cheers,
Christoph

[regina.lo...@gmail.com]

Hi Christoph,

many thanks for your suggestions. I will give it a try and will let everybody know if I arrive to any results. But first I need to solve my assembling problems (explained in another post).

Cheers,
Regina

[drur...@gmail.com]

Hi, I'm not sure if this is the appropriate place to ask a question, but I couldn't find anywhere else. I am trying to use MITObim with mira and had the same issue of using the wrong verion, 4.9.x. I am new to ubuntu and I've tried finding this on my own, but I can't seem to figure it out. How do I get mira version 4.0.2? I tried this

sudo apt-get purge mira-assembler

which seemed to remove it successfully, but when I tried to install 4.0.2, like this

sudo apt-get install mira-assembler=4.0.2

I received the error:

E: Version '4.0.2' for 'mira-assembler' was not found

I tried finding any extension that may need to go on that version, but I couldn't find anything that worked. Do you have any suggestion how to do this?

Thanks,

Austin

[Chris H]

Thanks for your patience! If you are running on ubuntu you can just download the precompiled binaries from here. Select the 'mira_4.0.2_linux-gnu..' in ubuntu. Download it. unpack the archive and you should be good to go. You can either add the binaries to your path or you tell MITObim where to find them via the '--mirapath' option.

cheers,
Christoph