SINEsFind

Author: Davide Carnevali (hackingfactory@gmail.com)

Abstract

Short Interspersed Element (SINE) retrotransposons are one of the most abundant DNA repeat elements in the human genome. They have been found to impact the expression of protein-coding genes, but the possible roles in cell physiology of their noncoding RNAs, generated by RNA polymerase (Pol) III, are just starting to be elucidated. For this reason, Short Interspersed Element (SINE) expression profiling is becoming mandatory to obtain a comprehensive picture of their regulatory roles. However, their repeated nature and frequent location within Pol II-transcribed genes represent a serious obstacle to the identification and quantification of genuine, Pol III-derived SINE transcripts at single-locus resolution on a genomic scale. Among the recent Next Generation Sequencing technologies, only RNA sequencing (RNA-Seq) holds the potential to solve these issues, even though both technical and biological matters need to be taken into account. A bioinformatic pipeline has been recently set up that, by exploiting RNA-seq features and knowledge of SINE transcription mechanisms, allows for easy identification and profiling of transcriptionally active genomic

How it works

SINEsFind allows to detect free SINE RNAs by using paired-end RNA-Seq data. It works by comparing the level of the expression coverage onto the SINE element with those upstream and downstream of it to it to check for enrichment. In this way SINEs_Find is able to distinguish a free PolIII-transcribed SINE RNA from one that is passenger of longer PolII transcript.

For a complete description of the method, check the research paper here

Requirements:

SINEs_find works with Python version 3.x and needs the following packages to be installed:

• Biopython
• EMBOSS suite
• HTSeq
• pyBedTools
• pyBigWig
• Pandas
• Numpy

Usage

SINEsFind works with paired-end RNA-Seq data, both stranded or not, but to exploit its potential it is recommended to use paired-end stranded reads at least 75 nt long. It works both with bam or bigwig files, the latter being much more faster (10x).

1. The first step is to build an index of the SINEs annotation by using the AnnoGenerate.py script which calculate the genomic coordinates of the expected full-lenght SINE element (see more in the reference paper below).
   It expects the following positional arguments:

• annotation: Annotation file in GTF format (either gzipped or not). Should refer to the same version of the human reference genome sequence
• genome: Human reference genome sequence. Should refer to the same version of the annotation file
• output: The output filename

Example: python AnnoGenerate.py annotation.gtf.gz hg38.fa outputname

The output file index will be the input GTF annotation file with additional 2 columns corresponding to the calculated start/end genomic coordinates of the expected full-length element. The annotation folder contains already the pre-computed extended annotation file for Alu elements in GRCh38 reference genome that do not overlap, in the same orientation, any of the genes annotated in Gencode version 24.

2. Once the extended annotation index has been created, we can run SINEsFind as in the following example which uses BAM file as input and default parameters:

python SINEs_find.py -s auto -t bam bamfile ./annotations/alu_intergenic_gencode24.gtf.gz chrom_sizes output_file

SINEs_find has several optional parameters that can be adjusted to user's need such as:

• The ratio of the expression coverage between the upstream/downstream regions and the SINE element
• The length of the upstream/downstream regions
• How many times the SINE expression coverage area should be greater than the calculated background

Please, run SINEs_find -h to view the Help and discover all available parameters and their default values, which are suitable for most of the cases.

Output

The output file of the SINEs_find tool contains the name and the genomic coordinates of all the putative SINE elements found. It also reports the expression coverage area of the upstream, central, downstream and 'out' regions, as well as the calculated background expression value for each chromosome.
When using BAM file as input, it save the expression coverage of the Watson and Crick strands inBedGraph format.

The expression coverage area is reported as number of bases covered by the reads mapping onto the corresponding region. Example: 10 paired-end reads of 100 nt lenght will be reported as 2000 (10x2x100)

Citations

If you use SINEs_Find in your research, please cite this paper

• Carnevali D, Dieci G. Identification of RNA Polymerase III-Transcribed SINEs at Single-Locus Resolution from RNA Sequencing Data. Non-Coding RNA 2017, 3(1), 15.