ECLIP data analysis using CTK

From Zhang Laboratory

Revision as of 10:25, 28 November 2020 by Czhang (Talk | contribs) (Demultiplexing samples)

Jump to: navigation, search
CTK home | Standard/BrdU-CLIP | iCLIP | eCLIP | PAR-CLIP | CTK usage | FAQ


This tutorial outlines how to analyze CLIP data derived from the enhanced CLIP (eCLIP) protocol. eCLIP incorporates modifications of the iCLIP protocol. Instead of utilizing circular ligation, the eCLIP protocol uses ligation of two separate adapters (an indexed 3' RNA adapter that is ligated to the crosslinked RNA fragment while on immunoprecipitation beads and and 3' ssDNA adapter that is ligated after reverse transcription) added in two separate steps. For more information on the biochemical iCLIP protocol, refer to the following reference:

Van Nostrand, E. L., G. A. Pratt, A. A. Shishkin, C. Gelboin-Burkhart, M. Y. Fang, B. Sundararaman, S. M. Blue, T. B. Nguyen, C. Surka, K. Elkins, R. Stanton, F. Rigo, M. Guttman and G. W. Yeo (2016). "Robust transcriptome-wide discovery of RNA-binding protein binding sites with enhanced CLIP (eCLIP)." Nat Methods 13(6): 508-514.

eCLIP reads typically have the following structure (read 2, which corresponds to the sense strand):


where N5 is the random barcode (UMI). In some libraries this could be 10 nt (N10).

S7 is the sample index, which at least the following possible 7mers: ATTGCTT (A01), ACAAGCC (B06), AACTTGT (C01), AGGACCA (D08). The sample index could also include 4 extra degenerate nucleotides, and takes the following form: ANNNNGGTCAT (A03), ANNNNACAGGA (G07), ANNNNAAGCTG (A04), ANNNNGTATCC (F05).

Sample dataset

We will use RBFOX2 eCLIP data from HepG2 cells for this tutorial, that were generated as part of the ENCODE project.

Van Nostrand, E. L., G. A. Pratt, A. A. Shishkin, C. Gelboin-Burkhart, M. Y. Fang, B. Sundararaman, S. M. Blue, T. B. Nguyen, C. Surka, K. Elkins, R. Stanton, F. Rigo, M. Guttman and G. W. Yeo (2016). "Robust transcriptome-wide discovery of RNA-binding protein binding sites with enhanced CLIP (eCLIP)." Nat Methods 13(6): 508-514.

Raw FASTQ files

We will use ~/test as our working directory, and download fastq files under ~/test/fastq

mkdir ~/test; cd ~/test
mkdir fastq; cd fastq

The raw sequence files from Illumina sequencing can be downloaded from the ENCODE web site.


#link files with more meaningful names 
ln -s ENCFF647KDW.fastq.gz HepG2.RBFOX2.rep1.R2.fastq.gz
ln -s ENCFF289OFA.fastq.gz HepG2.RBFOX2.rep2.R2.fastq.gz

Note that the read2 correspond to the sense strand. Following the original study (Van Nostrand et al. 2016), only read2 is used for analysis in this tutorial.

Output files of major steps generated in this tutorial

The number of reads in each file, after each step of processing, is summarized in Table 1 below.

Table 1: Summary of read numbers analyzed by CTK.
Protein Sample # of raw reads # of trimmed & filtered reads # of collapsed reads # of mapped reads # of reads after removal of repetitive RNAs # of unique tags
RBFOX2 rep1 10,038,858 9,907,629 7,652,372 4,809,641 4,498,170 4,094,510
RBFOX2 rep2 10,236,928 10,066,151 7,462,881 4,788,594 4,405,465 4,085,826
Total 20,275,786 19,973,780 15,115,253 9,598,235 8,903,635 8,180,336

Read preprocessing

Demultiplexing samples

The files downloaded from SRA have already be demultiplexed. Therefore this step has already been performed.

For the user who would like to start from a FASTQ file consisting of multiple libraries, check here.

In addition, in the demultiplexed fastq files, the random barcode has already been moved and attached to the begin of the read ID.

The barcode length can be either 5 nt or 10 nt, and it is important for this to be determined by checking a few reads, e.g.,

@CCAAC:SN1001:449:HGTN3ADXX:1:1101:1373:1964 2:N:0:1

Both replicates of RBFOX2 eCLIP have a barcode length of 5 nt. To be compatible with the CTK package, we reformatted the fastq files by moving the random barcode back to the beginning of each read (using a script not included in the CTK package as this is not a standard step of CLIP data analysis from raw data). The read illustrated above will become:

@SN1001:449:HGTN3ADXX:1:1101:1373:1964 2:N:0:1

Read quality filtering

For this particular dataset, we skipped the read quality filtering step, as this does not appear to make any major impact on the results.

For the user who would like to include this step, check here.

Trimming of 3' linker sequences

cd ~/test
mkdir filtering
cd filtering

One can then trim the 3' adapter using the command below.


for $f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
cutadapt -f fastq --match-read-wildcards --times 1 -e 0.1 -O 1 --quality-cutoff 5 -m 20 \
    -o $f.trim.fastq.gz ../fastq/$f.reformat.fastq.gz > $f.cutadpt.log

The parameters are as follows:

  • -f fastq : input is a FASTQ file
  • --match-read-wildcards : allow N in the read as matches
  • --times 1 : removes up to 1 adapters from each read; here we specify 2 as occasionally some tags have 2 copies of adapters
  • -e 0.1 : maximum allowed error rate is 0.1
  • -O 1 : the read is not modified if the overlap between the read and the adapter is shorter than 1nt
  • --quality-cutoff : refers to quality
  • -m 20 : minimum length of the read being 24nt (5 nt barcode + 15 nt tags)
  • -a : the 3' adapter

For more cutadapt usage information:

cutadapt --help

This is the same thing, but using a Sun Grid Engine to execute this UNIX job on a remote machine.

#$ -t 1-2 -m a -cwd -N CLIP

files=(HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2);

cutadapt -f fastq --match-read-wildcards --times 1 -e 0.1 -O 1 --quality-cutoff 5 -m 20 \
    -o $f.trim.fastq.gz ../fastq/$f.reformat.fastq.gz > $f.cutadpt.log

Note 1: We included all 8 version of possible 3' adaptors to be trimmed, as the exact adaptors used in the specific libraries were not documented.

Note 2: It is good to check the number of reads by running the following command:

#e.g., for raw reads
for f in `ls ../fastq/*.fastq.gz`; do
    c=`zcat $f | wc -l`
    echo $f $c

#e.g., for trimmed reads
for f in `ls *.trim.fastq.gz`; do
    c=`zcat $f | wc -l`
    echo $f $c

Collapse Exact PCR Duplicates

Collapse exact PCR duplicates such that if multiple reads have exactly the same sequence, only one is kept.

#!/bin/bash -l
for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2 HepG2.RBFOX2_mock_input.rep1.R2 
perl /usr/local/CTK/ $f.trim.fastq.gz - | gzip -c > $f.trim.c.fastq.gz

Note 1: It is good to check the number of reads by running the command:

for f in `ls *.trim.c.fastq.gz`; do
    c=`zcat $f | wc -l`
    echo $f $c

Strip random barcode (UMI)

Remove the 5' degenerate barcode. For these particular libraries, the barcode length is 5 nt (it could be 10 nt for some of the other eCLIP libraries).

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
    perl /usr/local/CTK/ -format fastq -len 5 $f.trim.c.fastq.gz - | gzip -c > $f.trim.c.tag.fastq.gz

Note 1: Get the distribution of tag lengths for diagnostic purposes using the following command.

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
    zcat $f.trim.c.tag.fastq.gz | awk '{if(NR%4==2) {print length($0)}}' | sort -n | uniq -c | awk '{print $2"\t"$1}' > $f.trim.c.tag.seqlen.stat.txt

Download fully preprocessed FASTQ files here(489 MB).

Read mapping & parsing

We assume the reference index has been prepared and is available under /genomes/hg19/bwa/.

cd ~/test
mkdir mapping
cd mapping

Run bwa to align the reads to the reference genome.

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
   bwa aln -t 4 -n 0.06 -q 20 /genomes/hg19/bwa/hg19.fa ../filtering/$f.trim.c.tag.fastq.gz > $f.sai
   bwa samse /genomes/hg19/bwa/hg19.fa $f.sai ../filtering/$f.trim.c.tag.fastq.gz | gzip -c > $f.sam.gz

The option -n 0.06 specifies slightly more stringent criteria than the default. The number of allowed mismatches (substitutions or indels) depending on read length is as follows:

[bwa_aln] 17bp reads: max_diff = 1
[bwa_aln] 20bp reads: max_diff = 2
[bwa_aln] 45bp reads: max_diff = 3
[bwa_aln] 73bp reads: max_diff = 4
[bwa_aln] 104bp reads: max_diff = 5
[bwa_aln] 137bp reads: max_diff = 6
[bwa_aln] 172bp reads: max_diff = 7
[bwa_aln] 208bp reads: max_diff = 8
[bwa_aln] 244bp reads: max_diff = 9

The -q option is used to trim low quality reads (an average of 20 or below). However, this does not seem to do too much after the trimming steps above.

Parsing SAM file

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
   perl /usr/local/CTK/ -v --map-qual 1 --min-len 18 --mutation-file $f.mutation.txt $f.sam.gz - | gzip -c > $f.tag.bed.gz

This will keep only unique mappings (with MAPQ >=1) and a minimal mapping size of 18 nt.

Note 1: In the tag bed file, the 5th column records the number of mismatches (substitutions) in each read

Note 2: Other aligners might not use a positive MAPQ as an indication of unique mapping.

Another useful option is --indel-to-end, which specifies the number of nucleotides towards the end from which indels should not be called (default=5 nt).

Note 3: The parsing script relies on MD tags, which is an optional field without strict definition in SAM file format specification. Some aligners might have slightly different format how they report mismatches. If other aligners than bwa is used, one should run the following command:

samtools view -bS $f.sam | samtools sort - $f.sorted
samtools fillmd  $f.sorted.bam /genomes/hg19/bwa/hg19.fa | gzip -c > $

This will ensure the sam file gets parsed properly.

Note 4: Keep track what proportion of reads can be mapped uniquely.

wc -l *.tag.bed

Remove tags from rRNA and other repetitive RNA (optional)

This is an optional step, and performed here following the original analysis of this dataset (see Van Nostrand et al. 2016 Nat Meth 13:508-514 for more detail). In some cases, CLIP data can have contaminants due to abundant repetitive RNAs (such as rRNAs).

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
   perl /usr/local/CTK/ -big -region /usr/local/CTK/annotation/genomes/hg19/annotation/rmsk.RNA.bed -ss --complete-overlap -r --keep-tag-name --keep-score -v $f.tag.bed $f.tag.norRNA.bed

As of now, the path to the BED file with a database of repetitive RNAs has to be specified directly, which is somewhat inconvenient. This will be improved in the future.

Collapse PCR duplicates

It is critical to collapse PCR duplicates, not only for the exact duplicates collapsed above, but also for those with slight differences due to sequencing errors. These sequencing errors can also occur in the random barcode, so a naive method to collapse the same barcode is frequently insufficient.

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
   perl /usr/local/CTK/ -big -v --random-barcode -EM 30 --seq-error-model alignment -weight --weight-in-name --keep-max-score --keep-tag-name $f.tag.norRNA.bed $f.tag.uniq.bed

In the command above, tags mapped to the same genomic positions (i.e., the same start coordinates for the 5' end of RNA tag) will be collapsed together if they share the same or "similar" barcodes. A model-based algorithm is used to identify "sufficiently distinct" barcodes. This algorithm was described in detail in the following paper:

Darnell JC, et al. FMRP Stalls Ribosomal Translocation on mRNAs Linked to Synaptic Function and Autism. Cell. 2011; 146:247–261.

Compared to the original algorithm, the current implementation estimates sequencing error from aligned reads, which is then fixed during the iterative EM procedure.

Note 1: Sequencing errors in the degenerate barcodes are estimated from results of read alignment. The number of substitutions in each read must be provided in the 5th column.

Note 2: Note that the read ID in the input BED file (in the 4th column) must take the form READ#x#NNNNN, where x is the number of exact duplicates and NNNNN is the bar-code nucleotide sequence appended to read IDs in previous steps. Read IDs that are not in this format will generate an error.

Note 3: As a diagnostic step, get the length distribution of unique tags, which should be a more faithful representation of the library:

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
   awk '{print $3-$2}' $f.tag.uniq.bed | sort -n | uniq -c | awk '{print $2"\t"$1}' > $f.tag.uniq.len.dist.txt

Get the mutations in unique tags.

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
   python /usr/local/CTK/ $f.mutation.txt $f.tag.uniq.bed 4 4 N $f.tag.uniq.mutation.txt

This script is from galaxy, and included for your convenience. The parameters 4 4 indicate the columns in the two input file used to join the two, and N indicates that only paired rows should be printed.

Table 2 summarizes the number of unique mutations of different types in each sample.

Table 2: Summary of mutations in unique tags.
Protein Sample # of unique tags Total # of mutations in unique tags Deletions Insertions Substitutions
RBFOX2 rep1 4,094,510 754,097 123,895 12,833 617,369
RBFOX2 rep2 4,085,826 616,218 72,627 11,418 532,173
Total 8,180,336 1,370,315 196,522 24,251 1,149,542

Merging biological replicates

After getting unique tags, one might want to concatenate these runs, which can be distinguished by different colors. As an example:

perl /usr/local/CTK/ -v -col "188,0,0" HepG2.RBFOX2.rep1.R2.tag.uniq.bed HepG2.RBFOX2.rep1.R2.tag.uniq.rgb.bed
perl /usr/local/CTK/ -v -col "128,0,128" HepG2.RBFOX2.rep2.R2.tag.uniq.bed HepG2.RBFOX2.rep2.R2.tag.uniq.rgb.bed

Then these can be concatenated.

cat HepG2.RBFOX2.rep1.R2.tag.uniq.rgb.bed HepG2.RBFOX2.rep2.R2.tag.uniq.rgb.bed > HepG2.RBFOX2.pool.R2.tag.uniq.rgb.bed
cat HepG2.RBFOX2.rep1.R2.tag.uniq.mutation.txt HepG2.RBFOX2.rep2.R2.tag.uniq.mutation.txt  > HepG2.pool.R2.tag.uniq.mutation.txt

Download unique tag and mutation files here (424MB).

Annotating and visualizing CLIP tags

Get the genomic distribution of CLIP tags

perl /usr/local/CTK/ -dbkey hg19 -ss -big -region -v -summary HepG2.RBFOX2.pool.R2.tag.uniq.annot.summary.txt HepG2.RBFOX2.pool.R2.tag.uniq.rgb.bed HepG2.RBFOX2.pool.R2.tag.uniq.annot.txt

Make sure the current genome (hg19) is specified. Check the summary file (HepG2.RBFOX2.pool.R2.tag.uniq.annot.summary.txt) for the percentage of tags mapped to CDS, 3'UTR, introns, etc.

Generate bedgraph for visualization in the genome browser

perl /usr/local/CTK/ -v -ss -exact -of bedgraph -n ″Unique Tag Profile″ HepG2.RBFOX2.pool.R2.tag.noRNA.uniq.rgb.bed HepG2.RBFOX2.pool.R2.tag.uniq.bedgraph

The output bedgraph file can be loaded into a genome browser for visualization of tag number at each genomic position.

Peak calling

cd ~/test
mkdir cluster
cd cluster
ln -s ../mapping/HepG2.RBFOX2.pool.R2.tag.uniq.rgb.bed

Mode 1: Peak calling with no statistical significance

perl /usr/local/CTK/ -big -ss -v --valley-seeking --valley-depth 0.9 HepG2.RBFOX2.pool.R2.tag.uniq.rgb.bed HepG2.RBFOX2.pool.R2.tag.uniq.peak.bed --out-boundary HepG2.RBFOX2.pool.R2.tag.uniq.peak.boundary.bed --out-half-PH HepG2.RBFOX2.pool.R2.tag.uniq.peak.halfPH.bed

Note 1: To annotate peaks with overlapping genes and repeat masked sequences and get genomic breakdown:

perl /usr/local/CTK/ -dbkey hg19 -ss -big -region -v -summary HepG2.RBFOX2.pool.R2.tag.uniq.peak.annot.summary.txt HepG2.RBFOX2.pool.R2.tag.uniq.peak.bed HepG2.RBFOX2.pool.R2.tag.uniq.peak.annot.txt

The output file HepG2.RBFOX2.pool.R2.tag.uniq.peak.annot.txt has the detailed annotation for each peak, and HepG2.RBFOX2.pool.R2.tag.uniq.peak.annot.summary.txt has summary statistics.

Note 2: Another useful option --valley-depth specifies the depth of the valley relative to the peak (0.5-1, 0.9 by default). One also has the option to merge peaks close to each other by specifying the distance between the peaks (e.g., -gap 20).

Note 3: Besides the peak boundaries (HepG2.RBFOX2.pool.R2.tag.uniq.peak.bed), one can also output cluster boundaries (--out-boundary) and half peak boundaries (--out-half-PH ) associated with each peak.

In this mode, the script will not assess the statistical significance of the peak height. If one needs this, an alternative mode is as follows:

Mode 2: Peak calling with statistical significance

perl /usr/local/CTK/ -big -ss -v --valley-seeking -p 0.05 --valley-depth 0.9 --multi-test --dbkey hg19 HepG2.RBFOX2.pool.R2.tag.uniq.rgb.bed HepG2.RBFOX2.pool.R2.tag.uniq.peak.sig.bed --out-boundary HepG2.RBFOX2.pool.R2.tag.uniq.peak.sig.boundary.bed --out-half-PH HepG2.RBFOX2.pool.R2.tag.uniq.peak.sig.halfPH.bed

Note 1: To get the number of significant peaks:

wc -l HepG2.RBFOX2.pool.R2.tag.noRNA.uniq.peak.sig.bed
39482 HepG2.RBFOX2.pool.R2.tag.uniq.peak.sig.bed

Note 2: To search for a known binding motif, one first defines the center of each peak (based on width at peak).

#extract from -500 to +500
awk '{print $1"\t"int(($2+$3)/2)-500"\t"int(($2+$3)/2)+500"\t"$4"\t"$5"\t"$6}' HepG2.RBFOX2.pool.R2.tag.uniq.peak.sig.bed >

This bed file can be used to extract sequences around peak (pay attention to the strand), and search for enrichment of specific motif (e.g., UGCAUG) relative to the peak.

Note 3: One might want to count the number of tags overlapping with each cluster/peak for each sample (e.g., to evaluate correlation between replicates).

for f in HepG2.RBFOX2.rep1.R2 HepG2.RBFOX2.rep2.R2; do
    perl /usr/local/CTK/ -ss -region HepG2.RBFOX2.pool.R2.tag.uniq.peak.sig.boundary.bed -of bed -v $f.tag.uniq.bed $f.tag.uniq.peak.sig.boundary.count.bed

Download output files for peak calling here (25 MB).


cd ~/test
mkdir CIMS

ln -s ../mapping/HepG2.RBFOX2.pool.R2.tag.uniq.rgb.bed 
ln -s ../mapping/HepG2.RBFOX2.pool.R2.tag.uniq.mutation.txt

Get specific types of mutations

Get specific types of mutations, such as deletions, substitutions, and insertions in unique CLIP tags

perl ~/czlab_src/CTK/ -t del HepG2.RBFOX2.pool.R2.tag.uniq.mutation.txt HepG2.RBFOX2.pool.R2.tag.uniq.del.bed
perl ~/czlab_src/CTK/ -t ins HepG2.RBFOX2.pool.R2.tag.uniq.mutation.txt HepG2.RBFOX2.pool.R2.tag.uniq.ins.bed
perl ~/czlab_src/CTK/ -t sub --summary HepG2.RBFOX2.pool.R2.tag.uniq.mutation.stat.txt HepG2.RBFOX2.pool.R2.tag.uniq.mutation.txt HepG2.RBFOX2.pool.R2.tag.uniq.sub.bed
#substitutions, as well as summary statistics of different types of mutations

It is always a good practice to look at the number of each type of mutation to see, e.g., compare the relative abundance of deletions to insertions. This information is provided in the summary statistics file when one specify the option --summary.


Here, we use deletions as an example.

perl /usr/local/CTK/ -big -n 10 -p -outp HepG2.RBFOX2.pool.R2.tag.uniq.del.pos.stat.txt -v HepG2.RBFOX2.pool.R2.tag.uniq.rgb.bed HepG2.RBFOX2.pool.R2.tag.uniq.del.bed HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.txt

By default, will output all sites including those that are not statistically significant. This is recommended because one can play with stringency later.

Extract sites that are statistically significant.

awk '{if($9<=0.001) {print $0}}' HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.txt | sort -k 9,9n -k 8,8nr -k 7,7n > HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.s30.txt
cut -f 1-6 HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.s30.txt > HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.s30.bed
wc -l HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.s30.bed
1403 HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.s30.bed

This will keep only those with FDR<0.001. Significant sites will also be sorted by FDR, then by the number of tags with mutations, and then by the total number of overlapping tags. One might try different thresholds to get a balance of sensitivity/specificity (e.g. as judged from motif enrichment).

Note 1: Another parameter that might be useful to improve signal to noise is m/k (i.g., $8/$7 in awk)

Note 2: By default, the command line above will analyze mutations of size 1. Sometimes deletion of multiple nucleotides occurs and those will be ignored here. One can analyze mutations of size 2 by specifying -w 2. Substitutions are always reported as a single nucleotide (even when consecutive nucleotides are substituted), and insertions occur technically in one position and are thus treated as size 1.

Note 3: The number of significant CIMS might be slightly different when unique tags are provided in different orders (this could occur easily when multiple replicates are concatenated together). This is because the FDR estimation is based on permutation, which can change slightly depending on the order of tags.

Note 4: To examine enrichment of motif around CIMS:

awk '{print $1"\t"$2-10"\t"$3+10"\t"$4"\t"$5"\t"$6}' HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.s30.bed > HepG2.RBFOX2.pool.R2.tag.uniq.del.CIMS.s30.21nt.bed
#+/-10 around CIMS

This BED file could be used to extract sequences around CIMS (pay attention to the strand), and search for enrichment of specific motif (e.g., UGCAUG for the case of RBFOX2) relative to the CIMS.

Download CIMS output files here (1.5 MB).

One can repeat these steps for the other types of mutations (i.e. substitutions and insertions). This is particularly relevant when one works with a new RBP, and it is unknown which type of mutations will be caused by cross linking.


cd ~/test
mkdir CITS
ln -s ../mapping/HepG2.RBFOX2.pool.R2.tag.uniq.rgb.bed
ln -s ../CIMS/HepG2.RBFOX2.pool.R2.tag.uniq.del.bed 
perl ~/czlab_src/CTK/ -big -p 0.001 --gap 25 -v HepG2.RBFOX2.pool.R2.tag.uniq.bed HepG2.RBFOX2.pool.R2.tag.uniq.del.bed HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.bed

In this script, we will look for reproducible CLIP tag start positions with more supporting tags than one would expect, which likely indicates crosslink induced truncation sites. For RBFOX2, since the vast majority of deletions are introduced because of cross linking, tags with deletions are treated as read-through tags and removed for CITS analysis.

Note 1: One can now perform motif enrichment analysis as described above in the CIMS section.

Note 2: In this particular example, we decided not to do Bonforroni correction, as we found this is too conservative. If necessary, this can be done by using option --multi-test.

Note 3: In the command line above, we opt to merge sites very close to each other and keep the most significant ones because CITS analysis tends to have more background when read through is frequent.

A few sites that span multiple nucleotides can be removed:

awk '{if($3-$2==1) {print $0}}' HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.bed > HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.singleton.bed
wc -l HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.bed HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.singleton.bed

14910 HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.bed
14823 HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.singleton.bed

Note 4: To examine enrichment of motif around CIMS:

awk '{print $1"\t"$2-10"\t"$3+10"\t"$4"\t"$5"\t"$6}' HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.singleton.bed > HepG2.RBFOX2.pool.R2.tag.uniq.clean.CITS.s30.singleton.21nt.bed
#+/-10 around CITS

This BED file could be used to extract sequences around CITS (pay attention to the strand), and search for enrichment of specific motif (e.g., UGCAUG for the case of Rbfox1-3) relative to the CITS.

Download CITS output files here (587 KB).

Motif enrichment

The enrichment of motif sites (when the motif is known) around CIMS or CITS provides a quantitative measure of the signal to noise ratio. The plot below shows the proportion of sites with the RBFOX2 binding UGCAUG motif starting at each position relative to CIMS/CITS (using tools not included in CTK). This analysis identifies G2 and G5 as the major crosslink sites between RBFOX2 and RNA.