riqc 0.2.0

Alignment-free RNA degradation tool

=== riqc ===

Alignment-free RNA degradation tool

• Free software: MIT license

Installation

The simplest is to install via:

pip install riqc

If you downloaded the code directly, you can also run:

python setup.py install

Usage

Degradation Detection in Aligned Files

Example:

riqc –bam_dir=’…’ –anno_fn=’…’ –out_dir=’…’ –log=’…’

Degradation Detection in Non-Aligned Files

Example:

riqc –fastq_dir=’…’ –genome=’…’ –anno_fn=’…’ –out_dir=’…’ –out_fn=’…’ –pickle_all=’…’ –pickle_filt=’…’ –log=’…’

All Options for Degradation Detection

• Input options

  • –bam_dir : Directory of bam files where we get a degradation score for every single bam file and one for all (default=’-‘)

  • –bam_fn : Specifies a single bam file as input (default=’-‘)

  • –fastq_dir : Directory of fastq files that will be analysed as one (so for individual degradation scores one has to give a directory with only one file or file-pair) (default=’-‘)

  • –cnt_dir : Directory of pre-produced tab delimited count files (mainly for experimental purposes; the preprocessed file contains all necessary information from the annotation file as well as counts for first and last exon), default=’-’

  • –anno_fn : Path to annotation file (supported formats: gtf, gff, gff3) (default=’-‘)

  • –genome : Path to genome file (supported format: fasta) (default=’-‘)

  • –gene_list : File with gene-names to use (default=’-‘)

  • –separate_files_ON : Consider all input files individually (default=False)

  • sparse_bam_ON : Input BAM files are in sparse hdf5 format (default=False)

• Output Options

  • –out_dir : Directory to store output in (default=’.’)

  • –out_fn : Prefix for output files (default=’out’)

  • –anno_tmp_fn : Name of file for temporarily storing annotation information (if the name is ‘’ it will automatically be set (in libs/annotation.py) to reflect whether the protein-coding-genes-filter has been used and whether the legacy option was set to True) (default=’’)

  • –pickle_all : Name of pickle file for temporarily storing all kmers (if it is None, the name will automatically be set (in libs/kmer.py) to reflect the chosen length of kmers) (default=None)

  • –pickle_filt : Name of pickle file for temporarily storing filtered/cleaned kmers (if it is None, the name will automatically be set (in libs/kmer.py) to reflect the chosen length of kmers)’, (default=None)

• General Options

  • –quant : What type of quantification to use (options: rpkm,raw) (default=’raw’)

  • –pseudo_count_ON : Add Pseudocounts to ratio (to also consider genes where we have 0 count on at least one end) (default=False)

  • –length : Only consider the 25% longest (uq), 50% medium (mq), or 25% shortest (lq) genes (default=’uq’)

  • –score_on_bases_ON : Calculate degradation score not from last and first exon but from a certain amount of bases at beginning and end that can be set via –base_number (normalized length) (default=False)

  • –base_number : Number of bases at beginning/end for calculating score (only relevant if –score_on_bases_ON is set) (default=100)

  • –log : Name of log file (default=’out.log’)

  • –verbose_ON : Set logger to verbose (default=False)

  • –plot_ON : Plot figures (default=False)

  • –fn_sample_ratio : Sample Ratios in relation to yours (default=os.path.join(os.path.realpath(__file__).rsplit(‘/’, 1)[:-1][0], ‘data’, ‘sampleRatios/TCGA_sample_a_ratio_uq.tsv’))

  • –mask_filter : Mask all read-counts below this integer (default=’0’)

  • –protein_coding_filter_OFF’ : Only consider genes that are protein-coding (default=True)

  • –length_filter_OFF : Only consider genes of certain length (specified via –length) (default=True)

  • –save_counts_ON : Store the exon counts in .npy and .tsv files for later use (via –cnt_dir) (default=False)

  • –legacy : Switch on some legacy behavior (currently only targeting alternate calculation of transcript length in libs/annotation.py) (default=False)

• Kmer Options (only relevant if using fastq as input)

  • –kmer_length : Length of k-mer for alignmentfree counting (default=27)

  • –reads_kmer : Required active reads per sample or if in [0, 1] then fraction of input reads considered (default=50000)

  • –step_k : Step-size for k-mer counting (default=4)

Additional Options for Degradation Compensation

• –scale_counts_ON : Scale counts with pre-computed scaling factors for degradation compensation (gi), default=False

• –scale_factors_dir : Directory of files containing scaling factors (default=’-‘)

The scaling-factor files can be generated with a command like

python …/degradation_tool/scalingFactors.py –bam_dir=’…’ –anno_fn=’…’ –out_dir=’…’

mainly using the same parameters as for the Degradation Detection with additional:

• –bins : Number of bins for different gene lengths (default=10)

• –relative_binning_ON : Have relative (to number of genes) bin boundaries instead of absolute values (default=False)

• –average_factors_ON : Compute scaling factors by using average (instead of median) per length bin (default=False)