Is there any protocol for whole genome sequencing data?

[y9c]

I wonder weather there is a example to apply cnvkit in whole genome sequencing data.

[etal]

Not here, since the files would be large. But I think there's an example in bcbio-nextgen to run it on the 1000 Genomes NA12878 data.

[y9c]

If I want to manipulate the data in na12878, which manual should I follow? I only found bed file in that folder.
Should I run make before I start these examples.

---

$make

......
Traceback (most recent call last):
  File "/usr/local/bin/cnvkit.py", line 4, in <module>
    __import__('pkg_resources').run_script('CNVkit==0.7.5.dev0', 'cnvkit.py')
  File "/usr/local/lib/python2.7/dist-packages/pkg_resources/__init__.py", line 745, in run_script
    self.require(requires)[0].run_script(script_name, ns)
  File "/usr/local/lib/python2.7/dist-packages/pkg_resources/__init__.py", line 1677, in run_script
    exec(script_code, namespace, namespace)
  File "/usr/local/lib/python2.7/dist-packages/CNVkit-0.7.5.dev0-py2.7.egg/EGG-INFO/scripts/cnvkit.py", line 11, in <module>

  File "build/bdist.linux-x86_64/egg/cnvlib/commands.py", line 576, in _cmd_fix

  File "build/bdist.linux-x86_64/egg/cnvlib/commands.py", line 589, in do_fix

  File "build/bdist.linux-x86_64/egg/cnvlib/fix.py", line 25, in load_adjust_coverages
  File "build/bdist.linux-x86_64/egg/cnvlib/fix.py", line 89, in match_ref_to_probes
ValueError: Reference is missing 19402 bins found in CL_seq
make: *** [build/CL_seq_flat.cnr] Error 1

[etal]

The files in the na12878 in this repo are work in progress, and it's going to be an example for whole exome sequencing, not whole genome. But WGS analysis is fairly straightforward, and you don't need any special data to do the analysis from scratch:

1. Download the NA12878 whole-genome sequencing reads as a BAM file from your preferred source (e.g. ENA, Illumina).
2. Download the corresponding reference genome GRCh37, if you don't already have it. Scan it with cnvkit.py access to produce a BED file. This will be both the "access" and "target" BED file.
3. Create an empty file to use as the "antitarget" BED": touch empty.bed
4. Run CNVkit on this genome with the batch command, specifying the e.g. cnvkit.py batch NA12878.bam -n -f GRCh37.fasta -g access.g37.bed -t access.g37.bed -a empty.bed --annotate refFlat.txt --split --target-avg-size 1000
5. The segmentation by CBS will probably be noisy, so try repeating segmentation with a stricter p-value threshold: cnvkit.py segment NA12878.cnr -t 1e5

[etal]

Sure, @charliecurnin, you can create a reference .cnn from a pool of normal samples. See: - https://cnvkit.readthedocs.io/en/stable/pipeline.html#reference - https://cnvkit.readthedocs.io/en/stable/nonhybrid.html Using a pooled reference is not required, but generally it will improve your results.