Plotting of aligned sequencing reads, assembled contigs or pan-genome graphs in BAM/CRAM format and visualization of genomic variants.
Wally is available as a Bioconda package, as a pre-compiled static binary from the release page, as a singularity container SIF file or as a minimal Docker container. You can also build Wally from source using a recursive clone and make.
git clone --recursive https://github.com/tobiasrausch/wally.git
cd wally/
make all
Wally uses subcommands for different visualization modes.
Wally needs a sorted and indexed BAM/CRAM file and a reference genome. The output is a genomic alignment plot of the region specified on the command-line.
wally region -r chrA:35-80 -g <genome> <input.bam>
You can specify a plot name using another colon separator
wally region -r chrA:35-80:myplot -g <genome> <input.bam>
Most often you probably want to use a BED file with regions of interest and just execute wally in batch. The 4-th column specifies the plot name. For instance to plot variants from a VCF file you can use:
bcftools query -f "%CHROM\t%POS\n" <input.vcf.gz> | awk '{print $1"\t"($2-50)"\t"($2+50)"\tid"NR;}' > regions.bed
wally region -R regions.bed -g <genome> <input.bam>
Simple BED files are used to provide gene or other annotations. The BED file needs to be bgzipped and indexed via tabix (tabix -p bed input.bed.gz). The required columns are chromosome, start, end and an identifier which is displayed if there is sufficient space.
wally region -b bed/gencode.hg19.bed.gz -r chr17:7573900-7574000 -g <genome> <input.bam>
The chromosome names of the genome FASTA file, BAM file and BED annotation file need to match, "chr11" and "11" are not considered identical.
You can specify custom colors for each annotation record in the BED file using Hex color codes (column 5 of the BED file). For instance, an annotation file anno.bed with this content:
chr17 7350100 7350900 regRed 0xFF0000
chr17 7351100 7351900 regLime 0x00FF00
chr17 7352100 7352900 regBlue 0x0000FF
chr17 7353100 7353900 regTeal 0x008080
can be visualized using
bgzip anno.bed
tabix -p bed anno.bed.gz
wally region -b anno.bed.gz -r chr17:7350000-7354000 -g <genome> <input.bam>
You can include multiple BAM files in a plot such as a tumor genome and a matched control in cancer genomics.
wally region -r chr17:7573900-7574000 -g <genome> <tumor.bam> <control.bam>
In fact, wally can be used to create "wallpapers" of genomes which gave rise to the tool name. For instance, the below command can be used to create a full genome view of 96 samples of a full plate of SARS-CoV-2 genomes.
wally region -y 20480 -r NC_045512.2:1-30000 -g NC_045512.2.fa Plate*.bam
Wally allows concatenating images of different regions horizontally using the --split option. This can be used, for instance, to zoom into a specific variant. On the command line the regions need to be separated by , without spaces. As an example, you can zoom into the N501Y variant of the alpha SARS-CoV-2 lineage using
wally region -s 3 -x 2048 -r NC_045512.2:22000-24000,NC_045512.2:23000-23100,NC_045512.2:23050-23070 -g NC_045512.2.fa <input.bam>
You can split horizontally and vertically at the same time to view, for instance, a somatic inter-chromosomal translocation.
wally region -s 2 -r chrA:35-80,chrB:60-80 -g <genome> <tumor.bam> <control.bam>
If you specify the regions in a BED file using the -R option then the split parameter operates row-wise, e.g., for -s 3 row 1-3 of the BED file make up the first image, row 4-6 the second image, and so on.
With -p you can switch on the paired-end view.
wally region -p -r chrA:35-80 -g <genome> <input.bam>
The paired-end coloring highlights candidate structural variants supported by read1 (R1) and read2 (R2). Below is a mapping of delly's structural variant types to wally's paired-end coloring. For inter-chromosomal translocations I assumed for this illustration that R1 maps to chromosome A and R2 maps to chromosome B.
inversion-type paired-end, --R1--> --R2-->, INV:3to3inversion-type paired-end, <--R1-- <--R2--, INV:5to5deletion-type paired-end, --R1--> <--R2--, DEL:3to5duplication-type paired-end, <--R1-- --R2-->, DUP:5to3inter-chr paired-end, A:--R1--> B:--R2--> leads to --A--> <--B-- junction, BND:3to3inter-chr paired-end, A:<--R1-- B:<--R2-- leads to <--A-- --B--> junction, BND:5to5inter-chr paired-end, A:--R1--> B:<--R2-- leads to --A--> --B--> junction, BND:3to5inter-chr paired-end, A:<--R1-- B:--R2--> leads to <--A-- <--B-- junction, BND:5to3
For large and complex structural variants, wally supports split views (as explained above). For instance, for an inter-chromosomal translocation you probably want to use a 2-way horizontal split with a larger image size.
wally region -p -x 2048 -y 2048 -s 2 -r chrA:35-80,chrB:60-80 -g <genome> <tumor.bam> <control.bam>
To visualize genomic breakpoints it's also helpful to highlight clipped reads -c and supplementary alignments -u.
wally region -cup -x 2048 -y 2048 -s 2 -r chrA:35-80,chrB:60-80 -g <genome> <tumor.bam> <control.bam>
For long read alignments or mappings of entire contigs, one often wants to explore all matches of a long input sequence with respect to the reference to visualize, for instance, complex rearrangements. For a single read or contig, you need a sorted and indexed BAM file and the reference genome.
wally matches -r <read_name> -g <genome> <input.bam>
Forward matches are colored in blue, reverse matches in orange and the black line traces all alignment matches along the input read sequence. You can also plot multiple reads to inspect, for instance, alignment concordance or heterogeneity using a list of reads as an input file.
wally matches -R <read.lst> -g <genome> <input.bam>
For many reads, this quickly leads to "wall(y)papers". With -s you get a separate plot for each read.
wally matches -s -R <read.lst> -g <genome> <input.bam>
You can also compare assemblies or align contigs to a reference genome using a workflow like this:
minimap2 -ax asm5 ref.fa asm.fa | samtools sort -o assembly.bam -
samtools index assembly.bam
wally matches -r <contig_name> -g ref.fa assembly.bam
All pairwise dotplots of a FASTA file can be generated using
wally dotplot sequences.fa
You can also extract reads directly from the BAM file and generate a reference-independent dotplot that includes the genomic mapping locations of read segments.
wally dotplot -R reads -g hg38.fa input.bam
For instance, to inspect haplotype differences or tumor heterogeneity at a given genomic locus (e.g., chrA:1000-2000) one can use:
samtools view input.bam chrA:1000-2000 | cut -f 1 | sort | uniq > reads
wally dotplot -R reads -g hg38.fa input.bam
Instead of plotting reads in pairs against each other, you can also calculate dot plots with respect to reference regions.
wally dotplot -g hg38.fa -R reads -e chrA:35000-80000 input.bam
wally dotplot -g hg38.fa -e chrA:35000-80000 sequences.fa
A minimum contig or read length can be specified using -s
wally dotplot -s 10000 -R reads -g hg38.fa input.bam
Lastly, you can flatten the genomic mappings into simple blocks using -f
wally dotplot -f -s 10000 -R reads -g hg38.fa input.bam
To visualize a self-alignment of a read with reference mappings use -a
wally dotplot -a -r read_name -g hg38.fa input.bam
You can try wally on some selected data sets using the wally web application.
Tobias Rausch, Rene Snajder, Adrien Leger, Milena Simovic, Mădălina Giurgiu, Laura Villacorta, Anton G. Henssen, Stefan Fröhling, Oliver Stegle, Ewan Birney, Marc Jan Bonder, Aurelie Ernst, Jan O. Korbel
Long-read sequencing of diagnosis and post-therapy medulloblastoma reveals complex rearrangement patterns and epigenetic signatures
Cell Genomics, 2023, 100281, DOI: 10.1016/j.xgen.2023.100281
Wally is distributed under the BSD 3-Clause license. Consult the accompanying LICENSE file for more details.
Wally relies heavily on the HTSlib and OpenCV. The visualization of genomic alignments was heavily inspired by IGV.