Intron Splicing Order

http://intron-splicing-order.online:3838/iso/

Prerequisite

R and R packages

packages within R

install.packages(c("readr","Rcpp","dplyr","igraph","dbscan","stringr","gtools","rstudioapi","gridExtra") )

BiocManager::install("lpsymphony")

Calling intron splicing order from users' own BAM files, then need install JAVA (JRE or JDK)

Oracle JDK8/JRE8

Steps

1. Aligning FASTQ reads with splice-wise aligner.

STAR, minimap2, et.al, then index the bam file

samtools index <Bam file>

2. Calculating intron splicing order pairs using the custome java program

java -jar java/isoLarge.jar -i anno/hg19_gencode_from_ucsc.bed -ibam <bam_file> -o <output_file> -t <optional INT e.g. 90>

The last parameter is the minium length of nucleotides aligned in intron side of intron-exon junction

Please put the output file under data/, since the R code will treat data/ as directory of intron splicing order pairs files.

Output format

Column	Meaning
Column #1	Transcript id
Column #2	Intron 1, the coordinate of relatively slower spliced intron
Column #3	Intron 2, the coordinate of relatively faster spliced introns (also include detected junctions)
Column #4	Strand
Column #5	Deprecated
Column #6	Read count supports this intron splicing order pair (intron 1 spliced after intron 2)
Column #7	Read count supports both two introns were spliced

3. Calculating most likely intron splicing orders

If users are not working with Rstudio, then will need to edit the run.R to change the working dir to intron_order

Source the below R script in Rstudio.

intron_order/code/run_human.R

For other genomes that are not included here

Prepare transcription ID and gene symbol information.

This can be easily got from ENSEMBL BioMart server, please the below file for an example.

data/hg19_ensembl_gene_id_trans_id_map.tsv

Column names are:

gene_id,trans_id,gene_symbol

Suggestions and comments are welcome: limeng49631@aliyun.com or use issues