As a demonstration of ChloroExtractors capabilities we assemble the spinach (Spinacia oleracea) chloroplast from whole genome shotgun sequencing data. We use the dataset SRR4447192 at SRA which consists of 301 bp paired Illumina reads. The raw reads were originally used to perform genome assembly of Spinacia oleracea to identify genomic SSRs. This demo can be executed in ~15min on any current desktop workstation or laptop with at least 8 GB of storage. I expect that ChloroExtractor and all its requirements are installed, otherwise please refer to the README.
If you have sratoolkit installed you can download the first 2 million reads of the dataset with the following command:
fastq-dump --split-files -X 2000000 SRR4447192
To run ChloroExtractor with default configuration just execute (replace with the path where ChloroExtractor is installed):
<chloroExtractor>/bin/ptx -1 SRR4447192_1.fastq -2 SRR4447192_2.fastq -o SRR4447192 -d SRR4447192
After approximately 10 minutes the ChloroExtractor should be done.
The final candidate chloroplast sequence is in SRR4447192/fcg.fa.
If everything worked this should be a single sequence with 151 kbp.
TODO: Discuss intermediate output, especially ptx.pdf, but also log
You might want to compare this chloroplast to other existing ones. For example the reference spinach chloroplast (NC_002202) and that of Arabidopsis thaliana (NC_000932) can be used (download the sequences as fasta from NCBI). One possibility to do this is using AliTV. If you have the perl part and requirements installed you can simply do:
<AliTV-perl-interface>/bin/alitv.pl Spinacia_oleracea_NC_002202.fasta Arabidopsis_thaliana_NC_000932.fasta fcg.fa
The resulting json file can be imported into this web interface. After reverse complementing and rotating (it is circular) the generated sequence, the whole genome alignment looks like this:
So our new sequence actually looks quite similar to the reference chloroplast.
TODO: Point to DOGMA and Verdant