Step 1 eDiVA-Predict

This step describes how to launch eDiVA predict on all new samples coming out of a sequencing machine.

nextflow run eDiVA-Predict.nf \
                    --NAME sample_name \
                    --READ1 fastq.read1.gz \
                    --READ2 fastq.read2.gz \
                    --AFFECTED 1
                    --CPU 2 \
                    --OUTF output_folder/ \
                    -w work_folder/  

Parameters

• NAME : Sample name you want to be in the Bam and VCF files
• READ1 and READ2 : fastq files with the raw reads
• AFFECTED : set to 1 if the sample is affected by the disease. This is optional and default is 0
• CPU : number of CPUs assinged to the task, 2 is the default
• OUTF: output folder where to put the Bam and VCF file. One foder per sample is recommended since the final output has alwasy the same name 'all_variants.vcf'
• -w : NextFlow work directory. It is important if you specify your own one for the manual cleanup of NextFlow or for the resuming of interrupted jobs

Outputs For each sample you will obtain:

• Bam file of aligned and filtered reads
• all_variants.vcf with the SNPs and INDELs calls
• sample_info.txt [required for the next steps]
• Quality_control folder with the fastqc results and other quality metrics for the sample.

---

Docker execution

Provided you have built ediva:code as an image and edited the nextflow.config as suggested in README from Docker folder, the execution of eDiVA-Predict is as follows: Remember that with Docker you can specify the amount of resources for each container, so edit the nextflow.config file if you want to add CPUs etc.

nextflow run eDiVA-Predict.nf \
                    --NAME sample_name \
                    --READ1 fastq.read1.gz \
                    --READ2 fastq.read2.gz \
                    --AFFECTED 1
                    --CPU 2 \
                    --OUTF output_folder/ \
                    -w work_folder/  \
                    -with_docker ediva:code 

---

Step 1.5 eDiVA_prepare_for_annotation