Nanopore - practical

• de.NBI nanopore training course
• a few example workflows
• in general always check the --help for each program

0. Base calling

• using Albacore
• Comparison of Basecaller here

# part of my script so variables are assigned to the options
read_fast5_basecaller.py -i $currDir/$i -f $flowcell -t $CPU -q 100000 -o fastq -k $kittype -r -s $currDir/FASTQ/

1. QC of reads

• Using the assembly-stats to get a summary of the reads:

assembly-stats ERR1147227.fastq

2. Adapter removal & demultiplexing

• with porechop
• porechop needs seperate outputs depending if barcodes were present or not

porechop -i <input>.fastq -b <output_foulder>

• You can redo assembly-stats to validate

3. Assembly

• with minimap2 and miniasm
• use canu for de novo assemblies
• minimap2 creates a files that miniasm needs for assembly
• minimap2 creates a first draft map. However its not a good minion only assembler, we use it here to combine it with illumina data

minimap2 -x ava-ont ERR1147227_trimmed.fastq ERR1147227_trimmed.fastq | gzip -1 > ERR1147227.paf.gz

miniasm -f ERR1147227_trimmed.fastq ERR1147227.paf.gz > ERR1147227.gfa

• canu example

canu -d run_e.coli -p e.coli genomeSize=6m -nanopore-raw 7718_E.coli_sum.fastq

4. Polishing

• We do: index build, mapping reads to assembly, piping to samtools for bam conversion, sorting, indexing
• -p4 for 4 threads (CPU) as bowtie2 takes time

bowtie2-build ERR1147227.fasta ERR1147227

bowtie2 -p4 -x ERR1147227 -1 ecoli_hiseq_R1.fastq.gz -2 ecoli_hiseq_R2.fastq.gz | samtools view -bS -o ERR1147227.bam

samtools sort ERR1147227.bam -o ERR1147227.sorted.bam

samtools index ERR1147227.sorted.bam

• Now we can do the actual polishing with pilon

pilon --genome ERR1147227.fasta --frags ERR1147227.sorted.bam --output ERR1147227_improved