PlaTypUS

A useful pipeline, but it takes effort to get it running. Victoria Corey has been very generous with her support. A few comments for setting up and running the pipeline on a Macbook Pro: 1. INSTALLATION: The readme says to to run install.sh, but this doesn't work! You must run platypus_install.sh instead. 2. The platypus version I downloaded in October, 2015 came with a version of GATK that does not work with my java (1.8...). I installed a new version of GATK on my machine (easy) and did this inside the platypus.sh script: #gatk_path=${working_dir}/bin/gatk/GenomeAnalysisTK.jar gatk_path=/Applications/GATK/GenomeAnalysisTK.jar 3. The default save location is your Desktop folder. I changed this as follows: #results_dir=~/Desktop/${new_name}"_"${start_time}/ results_dir=${working_dir}/results/${new_name}"_"${start_time}/ (This would be necessary for running it on something other than a Mac.) 4. TESTING: To test the Platypus I got some Pfal data from the SRA: GET DATA: fastq-dump --split-spot --split-files SRR1619351 (This is DD2, it would have been better if I had used 3D7) PREPARING TEST DATA: I aligned this like so: ALIGN: /path/bwa mem -M -R '@RG\tID:group1\tSM:sample1\tPL:illumina\tLB:lib1\tPU:unit1' \ /path/p_fal.fasta \ SRR1619351_1.fastq SRR1619351_2.fastq > p_fal_SRR1619351_p.sam Convert this to bam, and sort and index before using it with Platypus. (You may use the versions of bwa and samtools that come with Platypus, or you may use your own.) Testing succeeded! 5. USING PLATYPUS WITH A NEW SPECIES: I am running the Platypus on P.knowlesi. I got the most recent version of the P.knowlesi genome and annotations from plasmodb in November, 2015, and put them in the reference directory. The genome must be indexed for both samtools and bwa... samtools index genome bwa index genome To make the *.dict file: java -jar ${picard}/CreateSequenceDictionary.jar R=<filename.fasta> O=<newfile.dict> This does not produce the chromlist and genelist files. Model these on p_fal.chromlist and p_fal.genelist. If you are familiar with unix it is easy to make these files. The pipeline runs without a *.genome file. 6. To be continued...