cmap file in bioscope

[DNAjunk]

Hello

Just a simple question about where to put the cmap file for the mapping pipeline in bioscope 1.0:

First I put the self-made cmap file "my.cmap" into the $CORONAROOT/etc/cmap directory, but I get the error

PluginRunner:166 - java.io.IOException:~ReferenceFile~does~not~exist:~/data/scratch2/BIOSCOPE/sample/temp.02/my.cmap

the temp.02 was created by bioscope but indeed , there is no cmap file in it.

Any quick reply is appreciated.

PS: sometime I also get the errors

FATAL [main] PipelineExecutionHelper:143 - mapreads: FastaFile::Read(); ERROR: was expecting '>'
[mapreads] - Starting map 1 of 14 with pattern: 00110100011011100110001001011

and

FATAL [main] PluginRunner:166 - com.apldbio.aga.analysis.common.ExitException:~mapreads:~FastaFile::Read();~~
ERROR:~was~expecting~'>'
[mapreads]~-~Starting~map~1~of~14~with~pattern:~00110100011011100110001001011

What is the reason?

[westerman]

The cmap location has to be specified in the *.ini files. E.g.,

cmap = ${reference.dir}/Chicken/chicken_all.cmap

In other words the cmap does not reside in a default directory (that I am aware of). My 'reference.dir' that I use the in the example above is my common reference directory and not a bioscope one.

[DNAjunk]

ive defined cmap in my ini file.

how do you define the reference variable? comment out?

[westerman]

First I put the self-made cmap file "my.cmap" into the $CORONAROOT/etc/cmap directory,

Ah. Your original post (above) did not indicate that you defined where the cmap file was located. Instead it indicated that you had simply placed the cmap in a location and expected it to work from there.

As for the reference, it too has to have its path defined. In my global ini file I have:

reference.dir = /scratch/genomics/bioscope_references/

And in my per-project ini file I have the two lines:

reference = ${reference.dir}/Chicken/chicken_all.fasta
cmap = ${reference.dir}/Chicken/chicken_all.cmap

Thus both the reference and the cmap file locations are explicitly defined.

If you want to send me your *.ini I will take a look at them in order to see if I can spot anything obviously amiss with them.

[DNAjunk]

Thanks for your help. I'm sorry that I couldn't respond yesterday, but accessing seqanwers website didn't work for me... I have to admit that I am using bioscope (1.2) for the first time...

I have changed the ini files accordingly. However, the reference fileS confuse me... I have got 21 fasta files, one for each chromosome. In your case, did you concatenate the chicken references into one file chicken_all.fasta ? If so, is the cmap file still needed?

Here are my ini files: both are in the same directory where the ./run.sh script is started
1) global.ini:
############################
##
## global parameters
##
base.dir=./
output.dir = ${base.dir}/outputs
temp.dir = /data/scratch/temp
intermediate.dir = ${base.dir}/intermediate
log.dir = ${base.dir}/log
reads.result.dir.1 = ${base.dir}
reads.result.dir.2 = ${base.dir}
reference.dir = /data/bioseq02/cynomolgus/coronaValidated/
scratch.dir=/data/scratch

EOF

2) map.ini (my.ini)
##################################
##
## global settings for the pipeline run
##
import ./global.ini
cmap = ${reference.dir}/cyno.cmap
reference=${reference.dir}/chr1.fa
read.length = 27

##################################
##
## mapping pipeline
##
mapping.run = 1
mapping.tagfiles.dir =./reads
mapping.output.dir = ${output.dir}/chr
mismatch.level=2

##################################
##
## temp files and folders keep
##
#pipeline.cleanup.middle.files = 0
#job.cleanup.temp.files = 0

EOF

Remark: in the map.ini: if I don't give the name of one fasta file then bioscope doesn't not run reference=/data/bioseq02/cynomolgus/coronaValidated/chr1.fa . But in fact, I want that bioscope does the mapping on all chromosomes specified in the cmap file that looks like:

#DH10B w/ introduced indels
#chr_ID(must be digit)<TAB>chr_name<TAB>FASTA reference(must exist)<TAB>Double-Encoded Reference<NEWLINE>
1 chr1 /data/bioseq02/cynomolgus/coronaValidated/chr1.fa none
2 chr2 /data/bioseq02/cynomolgus/coronaValidated/chr2.fa none
3 chr3 /data/bioseq02/cynomolgus/coronaValidated/chr3.fa none
etc.

[westerman]

I have changed the ini files accordingly. However, the reference fileS confuse me... I have got 21 fasta files, one for each chromosome. In your case, did you concatenate the chicken references into one file chicken_all.fasta ? If so, is the cmap file still needed?

Yes I concatenated all of the chicken chromosomes into one file. Doing this makes the cmap file even more required. How else is the program going to be able to figure out which sequence in the concatenated file belongs to which file?

Do make sure that the entries in the cmap file correspond in order to the sequences in the concatenated file. A simple UNIX 'cat' command can mess you up since the sort order that is used may put 10 right behind 1 instead of the expected 2.

My chicken cmap looks like, in part:

1 1 /scratch/genomics/bioscope_references/Chicken/chr_1_NC_006088.fasta
2 2 /scratch/genomics/bioscope_references/Chicken/chr_2_NC_006089.fasta
...
9 9 /scratch/genomics/bioscope_references/Chicken/chr_9_NC_006096.fasta
10 10 /scratch/genomics/bioscope_references/Chicken/chr_10_NC_006097.fasta
11 11 /scratch/genomics/bioscope_references/Chicken/chr_11_NC_006098.fasta
...
...
29 32 /scratch/genomics/bioscope_references/Chicken/chr_32_NC_006119.fasta
30 W /scratch/genomics/bioscope_references/Chicken/chr_W_NC_006126.fasta
31 Z /scratch/genomics/bioscope_references/Chicken/chr_Z_NC_006127.fasta
32 mito /scratch/genomics/bioscope_references/Chicken/mito_NC_001323.fasta

The header lines from my concatenated file (via a "grep '>'" command):
>gi|118135635|ref|NC_006088.2|NC_006088 Gallus gallus chromosome 1, reference assembly (based on Gallus_gallus-2.1)
>gi|118135651|ref|NC_006089.2|NC_006089 Gallus gallus chromosome 2, reference assembly (based on Gallus_gallus-2.1)
...
and so on. In other words the same order as the cmap file.

[DNAjunk]

Many thanks for the detailed explanation.
I am using the mapping tool (command line version) only.
According to AB there is no cmap file needed for that.
I just concatenate all chromosomes into one file ( I didn't understand how the unix command cat can mess it up though...)
The usage of bioscope is not so easy, although there is a large bioscope manual. If there is anybody who has some kind of tutorial available, I would be happy to get a copy.

[westerman]

Yes, cmap is not needed for the mapping process. A cmap is needed for the InDel and I believe Inversion and SNP calling processes. Since I tend to follow up mapping with all three of the extra analyses I just always use a cmap file.

I agree that command line Bioscope is hard to use. GUI interface may be better.