kyber

Tool to quickly make a minimalistic 600x600 pixels heatmap image of read length (log-transformed) and read accuracy.

EXAMPLES

Gap-compressed percent identity	Phred-scale accuracy	Comparison of two runs

INSTALLATION

Download the appropriate binary from the releases.

USAGE

Usage: kyber [OPTIONS] <INPUT>

Arguments:
  <INPUT>  cram or bam file, or use `-` to read from stdin

Options:
  -t, --threads <THREADS>        Number of parallel decompression threads to use [default: 4]
  -o, --output <OUTPUT>          Output file name [default: accuracy_heatmap.png]
  -c, --color <COLOR>            Color used for heatmap [default: green] [possible values: red, green, blue, purple, yellow]
  -b, --background <BACKGROUND>  Color used for background [default: black] [possible values: black, white]
  -p, --phred                    Plot accuracy in phred scale
      --normalize                Normalize the counts in each bin with a log2
  -h, --help                     Print help
  -V, --version                  Print version

DETAILS

Both the x and y axis are fixed, allowing for comparison across datasets. The current settings should work for most (long-read) datasets, let me know if you disagree. The x-axis has log transformed read lengths, with a maximum length of 1M. The y-axis has the gap-compressed reference identity, ranging from 70% to 100%. When using Phred-scaled accuracy scores, the y-axis ranges from Q0 to Q40.

A 150 gigabase BAM file (from ONT PromethION) is processed in 11 minutes using 4 decompression threads (the default). If your input dataset is very large, you may want to consider to downsample it with samtools view -h and pipe that to kyber, e.g. samtools view -h -s 0.05 alignment.cram | kyber -

CITATION