SCRAM

SCRAM is lightweight Python package for aligning small RNA reads to one or more reference sequences and producing publication-quality images.

Developed by Stephen Fletcher @ the laboratory of Prof. Bernie Carroll, University of Queensland

Installation:

Scram is written in Python 3.5. Install scram and its dependencies with pip:

pip install scram

Or download and extract the tarball, then run:

python setup.py install

Input File Format:

Reference File : DNA nucleotides only (AGCT) - FASTA format

Sequence File : Collapsed reads - DNA nucleotides only (AGCT) - FASTA format

Post-processing of FASTQ reads to collapsed FASTA format can be carried out using the FASTX-Toolkit from the Hannon Lab. Collapsed reads are unique, and contain the read count in the header.

An example of the required read file format:

head seq1.fa

>1-607041
TCGGACCAGGCATCATTCCCC
>2-202886
TCGGACCAGGCTTCATACCCC
>3-71446
TCCCAAATATAGACAAAGCA

Usage:

scram analysis_type reference_file [-h] [-s1 SEQ_FILE_1] [-s2 SEQ_FILE_2] [-s3 SEQ_FILE_3] [-s4 SEQ_FILE_4] [-nt SRNA_LEN] [-f FILE_NAME] [-seq_list SEQ_LIST] [-min_read MIN_READ_SIZE] [-max_read MAX_READ_SIZE] [-min_count MIN_READ_COUNT] [-win SMOOTH_WIN_SIZE] [-ylim YLIM] [-no_display] [-split] [-pub] [-V]

Analysis types

• den : align reads of a single sRNA class (eg. 21 nt) from a single sequence file to a single reference sequence (-s1 and -nt required)
• mnt3dm : align 21, 22 and 24 nt reads from a single sequence file to a single reference sequence (-s1 required)
• CDP : count aligned reads of a single sRNA class (eg. 21 nt) to multiple reference sequences. Counts for two sequence files are plotted as (x,y) coordinates for each reference (-s1, -s2 and -nt required)

Flags

• -h : Help message
• -s1 : Sequence file/s 1 (if more than 1 file, read count is averaged for each read (if present in all files)
• -s2 : Sequence file/s 2 (if more than 1 file, read count is averaged for each read (if present in all files)
• -nt : sRNA length to analyse
• -f : Figure output file name (if not auto-generated)
• -p : No of cores (processors) to use (default=4)
• -min_read : Minimum length of sRNA reads used for normalisation (default=18)
• -max_read : Maximum length of sRNA reads used for normalisation (default=32)
• -min_count : Minimum read count for an sRNA to be aligned and used for normalisation (default=1)
• -win : Window size for smoothing of den plots (default=50)
• -ylim : +/- y-axis limit on plots
• -no_display : Do not display plot on screen
• -no_csv : Do not generate .csv alignment file
• -split : Split CDP read alignment counts based on no. of alignments
• -pub : Remove all labels from density maps for publication
• -bokeh : For Jupyter notebook inline plotting when scram started using magic run. No figure output to file.
• -V : Show program's version number and exit

den example:

scram den ./ref.fa -s1 seq1.fa -nt 24 -win 30 -f fig1.pdf

mnt3dm Example:

scram mnt3dm ./ref.fa -s1 seq1.fa -win 20 -ylim 110 -f fig2.pdf

CDP Example:

scram CDP ./cDNAs.fa -s1 seq1.fa -s2 seq2.fa -nt 21 -f fig3.pdf -split

(c) 2016 - Stephen Fletcher. MIT License