Giraffe-View 0.0.9.4

A small tool help assess and visualize the accuracy of a sequencing dataset, specifically for Oxford Nanopore Technologies (ONT) long-read sequencing.

Giraffe_View

Giraffe_View is designed to help assess and visualize the accuracy of a sequencing dataset, specifically for Oxford Nanopore Technologies (ONT) long-read sequencing including DNA and RNA data. There are four main functions to validate the read quality.

• observe calculates the observed read accuracy, mismatches porportion, and homopolymer identification.
• estimate calculates the estimated read accuracy, which is equal to Quality Score.
• gc_bias compares the relationship between GC content and read coverage.
• modi_bin perform statistics on the distribution of modification based on the bed file.

Install

To use this software, you need to install additional dependencies including samtools, minimap2, seqkit, and bedtools for read processing. The following commands can help you to install the package and dependencies.

pip install Giraffe-View
conda install -c bioconda -c conda-forge samtools minimap2 seqkit bedtools -y

General Usage

Giraffe View is run simply with fllowing commands:

giraffe -h
usage: giraffe [-h] {observe,estimate,gc_bias,modi_bin} ...

A tool to help you assess quality of ONT data.

positional arguments:
  {observe,estimate,gc_bias,modi_bin}
    observe             Observed quality in accuracy, mismatch, and homopolymer identification
    estimate            Estimated read accuracy
    gc_bias             Relationship between GC content and depth
    modi_bin            Average modification proportion of bins

optional arguments:
  -h, --help            show this help message and exit

The available sub-commands are:

observe

giraffe observe -h
usage: giraffe observe [-h] --input <fastq> --ref <reference> [--cpu <number>] [--plot]

optional arguments:
  -h, --help         show this help message and exit
  --input <fastq>    input reads
  --ref <reference>  input reference
  --cpu <number>     number of cpu (default:10)
  --plot             Results visualization

• fastq - the raw fastq data, some filter steps will be conducted including short read ( < 200 bp) and low quality read ( < 7 ) removal.
• reference - the reference file in fasta format.
• cpu - the number of CPUs will be used during processing.

estimate

giraffe estimate -h
usage: giraffe estimate [-h] --input <fastq> [--cpu <number>] [--plot]

optional arguments:
  -h, --help       show this help message and exit
  --input <fastq>  input reads
  --cpu <number>   number of cpu (default:10)
  --plot           Results visualization

gc_bias

giraffe gc_bias -h
usage: giraffe gc_bias [-h] --ref <reference> --input <sam/bam> [--binsize] [--plot]

optional arguments:
  -h, --help         show this help message and exit
  --ref <reference>  input reference file
  --input <sam/bam>  input bam/sam file
  --binsize          input bin size (default:1000)
  --plot             Results visualization

• reference - the reference file in fasta format.
• sam / bam - the result of mapping in sam/bam file. If you have used the observe function to process your data, the resulting tmp.sort.bam file can be used as the input.
• binsize - the length of bin. A bin is the smallest unit to count the read coverage and GC content.

modi_bin

giraffe modi_bin -h
usage: giraffe modi_bin [-h] --input <bed> --ref <reference> [--cpu <number>] [--plot]

optional arguments:
  -h, --help         show this help message and exit
  --input <bed>      input modificated bed file, please use the .bed as the file suffix
  --ref <reference>  input position file with CSV format, please use the .csv as the file suffix
  --cpu <number>     number of cpu (default:10)
  --plot             Results visualization

• bed - a BED file with four columns (three columns for position, one for methylation proportion). Please use the tab ("\t") to gap the column instead of the space (" ").

  #chrom	start	end	value
  chr1	81	83	0.8
  chr1	21314	21315	0.3
  chr1	32421	32422	0.85
  

• reference - a CSV file with target regions.

  chr1,0,100000,1_0_100000
  chr1,100000,200000,1_100000_200000
  

Workflow

graph TD
	A(raw signal) -.-> |Basecall| B(FASTA)
	A(raw signal) -.-> |Basecall| C(modificated BED)
	C(modificated BED) --> |modi_bin| D(modification distribution)
	B(FASTA) --> |estimate|e(estimated accuracy)
	B(FASTA) --> |observe| f(clean reads)
	f(clean reads) --> |observe| g(aligned BAM)
	
	g(aligned BAM) --> |observe|h(homopolymer identification)
 	g(aligned BAM) --> |observe|i(observed accuracy)
	g(aligned BAM) --> |gc_bias|j(GC bias) 

Developing

• run the homopolymer identification with multi-processed