winnowmap

Winnowmap Alignment

Align long-reads with Winnowmap.

From the reads, top 0.02% of the repetitive 15-mers were collected using Meryl and downweighted in the process of minimizer sampling. Use platform specific mapping parameters; map-pb for HiFi, map-pb-clr for CLR, and map-ont for ONT reads, respectively. Reads are sorted and indexed using samtools, and filtered for primary reads with samtools view -F0x100.

1. init.sh

# Collect repetitive 15-mers
meryl count k=15 $ref output merylDB
meryl print greater-than distinct=0.9998 merylDB > repetitive_k15.txt

2. map.sh

This step can be run in parallel per input read file. Use split.sh to split a file by every X num. lines (e.g. 100 M).

# Map and sort reads
winnowmap --MD -W repetitive_k15.txt -ax $map $opt -t$cpus $ref $reads > $out.sam
samtools sort -@$cpus -m2G -T $out.tmp -O bam -o $out.sort.bam $out.sam

3. merge.sh

# Merge and index
samtools merge -O bam -@$SLURM_CPUS_PER_TASK $prefix.bam *.sort.bam
samtools index $prefix.bam

4. filter.sh

# Filter to get primary read alignments
samtools view -F0x104 -@$SLURM_CPUS_PER_TASK -hb $in > $out
samtools index $out

5. sam2paf.sh

Finally, convert the desired bam file to paf using paftools from minimap2 and processe to coverage analysis. This script can be found under coverage.

./sam2paf.sh $prefix.pri.bam $prefix.pri.paf

Optionally, spurious alignment blocks shorter than 1 kb or lower than 85% identity can be removed from the ONT primary alignments.

# Filter ONT for alignment blocks < 1kb and identity < 85%
awk '$11>1000 && $10/$11>0.85' ont.pri.paf > ont.pri.len1k_idy85.paf

On Slurm

The entire job can be submitted using _submit.sh.

Usage: ./_submit.sh ref.fasta prefix map [wm_opt]
  Required: input.fofn

Collect all input reads to input.fofn, provide ref.fasta, prefix, and map option (map-pb / map-pb-clr / map-ont). Additional winnowmap options could be provided in wm_opt, with quotes.