NAME
====
REDO - RNA Editing Detection in Organelle
VERSION
=======
Version 1.0
COPYRIGHT
=========
Copyright 2016 - Wanfei Liu & Qiang Lin - Beijing Institute of Genomics, Chinese Academy of Sciences.
This tool is a free package; you can redistribute it and/or modify it freely.
INTRODUCTION
============
REDO is a comprehensive application tool for identifying RNA editing events in organelles based on variant call results (VCF files). It is a suite of Perl scripts and can work easily and directly in any operating system installed Perl and R Environment. The stringent rule depended filters and statistical filters are used in REDO for reducing false positive rate. It can provide detailed annotations, statistics and figures for RNA editing sites. REDO also can detect RNA editing events in multiple samples simultaneously and identify differential proportion of RNA editing events in different samples. Moreover, the genome variation can be easily removed by a subprogram fish.pl in our package.
REDO is available at https://sourceforge.net/projects/redo/.
REDO only uses the variant calling format (VCF) files (records for all sites), the genome sequence file (for annotation) and the gene annotation file (for annotation, feature table file, http://www.ncbi.nlm.nih.gov/projects/Sequin/table.html) as input, then the raw variants are filtered by rule-depended filters and statistics filters for reducing the false positive rate as following steps. 1. Quality control filter: the low quality sites are filtered according to the reads quality (DP=0 in GATK (reads with MQ=255 or with bad mates are filtered) and DP4=0 in samtools (number of high-quality ref-forward, ref-reverse, alt-forward and alt-reverse bases)). 2. Depth filter: three measures are used in this filter, which are total reads depth (<4), alt reads depth (<3), and the total reads depth variation between alt site and ad-jacent sites without variant in specific window (total reads depth<0.2*average reads depth of adjacent sites). 3. Alt proportion filter: two measures are used in this filter, including alt proportion (<0.1) and the reduced alt proportion after removing possible sequencing error proportion obtained from the adjacent sites without variant in specific window (alt proportion ¨C error proportion<0.1). 4. Multiple alt filter: only the variant with one alt allele is kept for RNA editing detection. 5. Distance filter: the variant sites with short distance from each other (<3) are filtered due to the possible position obstruction for RNA editing. 6. Splice junction filter: variants within short splice anchor (<2) are removed. 7. Indel filter: the indel variants are removed in default (indels can be kept when assigning "-i yes" option). 8. Likelihood ratio test filter: a likelihood ratio (LLR) test is used for RNA editing sites (Chepelev, 2012; Sun, et al., 2016). LLR test is a probabilistic test incorporating error probability of bases (error probability is obtained using adjacent sites without variant in specific window) for detecting RNA editing sites (LLR<10). 9. Fisher exact test filter: we assessed the significance for a given RNA editing site (alt reads, ref reads) by comparing its expected levels (0, alt reads + ref reads) using the Fisher¡¯s exact test and the p-value of fisher exact test is used for filter (p-value>0.01). 10. The complicated model filter: based on the statistics results for ex-periment validated RNA editing sites and the attributes of codon table, a complicated filter model was built according to five characteristics of RNA editing sites, which are RNA editing types (C->T, A->G, T->C, etc), alt proportion, amino acids change, codon phase and hydropho-bic/hydrophilic change.
PREREQUISITES
=============
Before running, the following software or program are required:
- Perl Environment (perl v5.18.4 or later (it can be lower), tested for Windows and Linux)
- Perl module: Getopt::Long; Text::NSP
- R Environment (tested on R 3.1.2 or later (it can be lower) for Windows and Linux)
- R module: graphics; grDevices
REDO have been tested on Windows and Linux.
INSTALLATION
============
This package doesn't need install. You can run it using absolute path after decompressed it. If you want to use it without absolute path, you should add the absolute path of REDO directory in PATH of your bash profile.
COMMAND LINE
============
Run REDO.pl without any parameter, it will print below usage on the screen.
*************
*1.0 version*
*************
************************************************************************
Usage: REDO.pl -g genome_sequence -v variant_file -t tbl_file -o prefix_of_outfile -d reads_depth -c minimum_coverage_of_alt_allele -p alt_proportion -w window_size -s minimum_splice_anchor -a minimum_alt_distance -l llr_value -f fisher_pvalue -dv depth_variant -i identify_indel.
-g: the absolute path of genome sequence.
-v: the absolute path of variant file/files (for example, "-v test1.vcf -v test2.vcf -v test3.vcf ......").
-t: the absolute path of tbl file (feature table file, http://www.ncbi.nlm.nih.gov/projects/Sequin/table.html).
-o: the prefix of output file.
-d: minimum reads depth (4).
-c: minimum coverage of alternative (alt) allele (3).
-p: minimum alt proportion (0.1).
-w: minimum window size for calculating error rate and average depth (10).
-s: minimum splice anchor size (2).
-a: minimum alt distance (3).
-l: minimum likelihood ratio (LLR) (10).
-f: maximum p-value of fisher exact test (0.01).
-dv: the depth variant (0.2).
-i: identify indel (y or n, default n).
************************************************************************
Note: all files should be uncompressed files. For VCF file, it should include all positions with reads supports (samtools/bcftools default VCF output file or GATK output file with ¡°-ERC BP_RESOLUTION¡±). We can provide one VCF file or multiple VCF files simultaneously. tbl file is the standard feature table file in NCBI (http://www.ncbi.nlm.nih.gov/projects/Sequin/table.html), we can download it from each genome sequence record by click ¡°Send -> Complete Record -> File -> Format -> Feature Table -> Create File¡± easily in nucleotide database of NCBI. If you want run REDO multiple times with different parameters, please use ¡°-o¡± option to assign different prefix name for output files. For calculating the error rate and the average reads depth, a default 10bp size window is used. To filter false positive sites, we use 4 as the default minimum reads depth and 3 as the minimum alt allele reads coverage, because most of false positive variants identified by GATK and samtools come from the low depth and alt coverage records according to our previous study. RNA editing may need second structure or motif to recognize related enzymes and we didn¡¯t find the RNA editing sites with short distance from each other in several experiment data except for Arabidopsis (<3bp). Therefore, we use 3 as the minimum alt distance and 2 as the minimum splice anchor size. According to the test using experiment data, the likelihood ratio (LLR) should be equal or larger than 10 (default value). Moreover, according to the statistics results for experiment validated RNA editing sites and the attributes of codon table, we applied a complicated filter model. The filter model based on five characteristics of RNA editing sites, which are RNA editing types (C->T, A->G, T->C, etc), alt proportion, amino acids change, codon phase and hydrophobic/hydrophilic change. Using this filter model, we lost 0%~1.34% sensitivity, but increased 0.96%~10.45% precision rate and 0%~0.13% specificity percent. Even though, our sensitivity is comparable with REDItools. Furthermore, the chromosome ID in genome sequence file should be identical with vcf and tbl file. At present, we only identify C->T RNA editing sites with at least 0.50 editing proportion in tRNA and rRNA region.
INPUTFILES
==========
REDO need three input files:
- VCF file (Variant Call Format, the result of GATK and samtools). REDO can accept one or multiple VCF files by option "-v".
- Reference genome in Fasta format. Chromosome/region names must be equal to those in VCF file and tble file.
- tbl file (feature table file, http://www.ncbi.nlm.nih.gov/projects/Sequin/table.html). It can be downloaded from NCBI directly.
Note: the chromosome ID in genome sequence file should be identical with vcf and tbl file.
OUTPUTFILES
===========
REDO mainly produces output files in simple tab-delimited table files.
1. *_sample.out
The RNA editing result for each sample.
Main fields are:
- Chr: chromosome name;
- Strand: strand information ("+" or "-");
- Name: gene name;
- Genome_pos: nucleotide position in chromosome (1-based);
- Gene_pos: nucleotide position in gene (1-based);
- AA_pos: amino acid position in gene (1-based);
- Phase: codon phase of editing site (1-based);
- Ref->Alt: reference allele to editing allele;
- RefCodon->AltCodon: reference codon to editing codon;
- RefAA->AltAA: reference amino acid to editing amino acid;
- AltRatio: editing reads proportion;
- ErrorRatio: sequencing error rate obtained from the adjacent sites without variant in specific window (default window size is 10);
- Depth: total reads number;
- AverageDepth: average total reads number of adjacent sites without variant in specific window (default window size is 10);
- RefRead: reference allele reads number;
- AltRead: editing allele reads number;
- AdjacentAlt: distance to adjacent candidate editing site;
- LLR: a likelihood ratio (LLR) is a probabilistic test incorporating error probability of bases (error probability is obtained using adjacent sites without variant in specific window) for detecting RNA editing sites;
- Pvalue(fisher): p-value is obtained by assessing the significance for a given RNA editing site (alt reads, ref reads) by comparing its expected levels (0, alt reads + ref reads) using the Fisher¡¯s exact test;
- RNAeditGC: GC content of RNA editing site in specific window (default 20bp);
- GeneGC: gene GC content;
- Exon_num: exon number of gene;
- Exon_start: exon start/starts of gene (multiple exon starts are separated by comma) (1-based);
- Exon_end: exon end/ends of gene (multiple exon ends are separated by comma) (1-based);
- Function: gene function.
2. *.out
The integrated RNA editing result for all samples.
Main fields are:
- Chr: chromosome name;
- Strand: strand information;
- Name: gene name;
- Genome_pos: nucleotide position in chromosome (1-based);
- Gene_pos: nucleotide position in gene (1-based);
- AA_pos: amino acid position in gene (1-based);
- Sample_num: total sample number of supporting RNA editing site;
- Phase: codon phase of editing site (1-based);
- Sample_name1: the detailed editing information in the sample, including Ref->Alt, RefCodon->AltCodon, RefAA->AltAA, AltRatio, ErrorRatio, Depth, AverageDepth, RefRead, AltRead, AdjacentAlt, LLR, Pvalue(fisher), RNAeditGC, and GeneGC separated by colon;
- Sample_name2;
- ......
- Exon_num: exon number of gene;
- Exon_start: exon start/starts of gene (multiple exon starts are separated by comma) (1-based);
- Exon_end: exon end/ends of gene (multiple exon ends are separated by comma) (1-based);
- Function: gene function.
3. *.change_matrix
The statistics result file for RNA editing in each sample, including total number and percent for each RNA editing type.
Main fields are:
- Prefix: prefix of sample name;
- Column 2 to column 13 are 12 possible RNA editing types, the total RNA editing number and percent for each type are shown.
4. *.stat
The statistics result file for RNA editing in each sample, including total editing number, indel editing number, editing number in different codon phases, silent number, non-silent number, and the detailed statistics for each RNA editing type.
Main fields are:
- Prefix: prefix of sample name;
- Total: total RNA editing number;
- Other: non-substitution (indel) RNA editing number;
- Phase(1,2,3): total RNA editing number in phase 1, 2 and 3 of codons for substitution RNA editing;
- Silent: silent RNA editing number for substitution RNA editing;
- NonSilent: non-silent RNA editing number for substitution RNA editing;
- A->C(silent): silent RNA editing number for A->C RNA editing;
- A->C(non_silent): non-silent RNA editing number for A->C RNA editing;
- A->C(1:hydrophobic2hydrophobic,2:hydrophilic2hydrophilic,3:hydrophobic2hydrophilic,4:hydrophilic2hydrophobic): the number of hydrophobic/hydrophilic changes due to A->C RNA editing in four types separated by comma;
- Column 10 to column 42 are RNA editing number in silent, non-silent and 4 types of hydrophobic/hydrophilic changes for other 11 RNA editing types.
5. *_sample.R and *_sample.jpeg
*_sample.R is the R script for drawing the attributes of RNA editing in each sample while *_sample.jpeg is the figure produced by *_sample.R, which shown 16 subfigures to illustrate the RNA editing attributes.
6. *_DPR.out
The pairwise sample comparison for differential proportion of RNA editing events.
Main fields are:
- #sample name1 vs sample name2: the sample name of comparison;
- Chr: chromosome name;
- Position: nucleotide position in chromosome (1-based);
- Proportion1: editing allele proportion in sample one;
- Alt1: editing allele reads in sample one;
- Ref1: reference allele reads in sample one;
- Proportion2: editing allele proportion in sample two;
- Alt2: editing allele reads in sample two;
- Ref2: reference allele reads in sample two;
- Fold(log2): log2 fold change comparing sample two to one;
- Pvalue: p-value is obtained by assessing the significance for a RNA editing site in sample one (alt1 reads, ref1 reads) by comparing to sample two (alt2 reads, ref2 reads) using the Fisher¡¯s exact test.
7. *_cluster.out, *_cluster.R and *_cluster.jpeg
*_cluster.out is a matrix for editing allele proportion of each RNA editing site in all samples. The columns are samples and the rows are RNA editing sites. *_cluster.R is the R script for doing the cluster analysis and *_cluster.jpeg is the cluster heatmap for RNA editing sites.
TEST & EXAMPLE
==============
REDO has been tested on coconut cp genome (KX028884) and related RNA-seq data (SRR1063404, SRR1063407, SRR1137438, SRR1173229, SRR1265939, SRR1273070, SRR1273180 and SRR606452, for comparing with REDItools) and Arabidopsis mt (NC_001284) and related RNA-seq data (SRR2079784, for comparing with REDItools). The cp RNA-seq data were mapped by GSNAP while mt RNA-seq data was mapped by BWA. All variants were identified by samtools. The small organelle genome, small VCF file and low memory request can make REDO run very fast in almost any computer (less than 2 minutes for 8 samples in coconut cp genome and less than 1 minute for 1 sample in Arabidopsis mt).
When you get the REDO package, decompress the directory, then you can test the program by following commands (just selected one command for each step).
1. Data pre-processing
1.1 Filter
java -jar trimmomatic-0.33.jar PE -threads 8 -trimlog SRR1063404.trimlog SRR1063404_1.fastq SRR1063404_2.fastq -baseout SRR1063404Filtered.fq.gz ILLUMINACLIP:TruSeq3-PE.fa:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:36&
1.2 Mapping
GSNAP:
gsnap -d coco_cp -a off -N 1 -A sam -t 8 --force-xs-dir --pairmax-rna=1000000 --split-output= SRR1063404_cp SRR1063404_1.paired.fq SRR1063404_2.paired.fq
BWA:
bwa mem -t 16 -M ath SRR2079784_1.paired.fastq.gz SRR2079784_2.paired.fastq.gz 1>SRR2079784.sam 2> SRR2079784.maperr &
1.3 Variant calling
Sam to bam: samtools view -b SRR2079784.sam -o SRR2079784.bam &
Sam sort: samtools sort -o SRR2079784.sorted.bam -O bam -T SRR2079784.temp SRR2079784.bam &
Sam mpileup: samtools mpileup -v -f arabidopsis_mt.fa SRR2079784.sorted.bam --output SRR2079784.sorted.vcf&
bcftools: nohup bcftools call -o SRR2079784.sorted.call.vcf -O v -c SRR2079784.sorted.vcf &
Note: we only used *.concordant_mult and *.concordant_uniq files as the final GSNAP mapping result while BWA default result was used as mapping result.
Table 1 The information of all test RNA-seq data
Species Organelle SRA accession No. Length Original fragments High quality fragments Percent Mapping fragment Percent
Coconut cp SRR1063404 202 36,009,632 32,555,041 90.41% 1,499,762 4.61%
SRR1063407 202 35,467,948 32,141,745 90.62% 101,308 0.32%
SRR1137438 152 50,839,994 42,267,444 83.14% 68,693 0.16%
SRR1173229 202 119,333,177 113,394,045 95.02% 249,845 0.22%
SRR1265939 202 51,540,183 48,892,847 94.86% 34,510 0.07%
SRR1273070 337 40,564,276 37,752,443 93.07% 10,508 0.03%
SRR1273180 252 60,030,680 54,291,251 90.44% 1,016,376 1.87%
SRR606452 180 27,465,703 27,063,513 98.54% 548,364 2.03%
Arabidopsis mt SRR2079784 132 21,839,182 20,471,554 93.74% 1,796,028 8.77%
2. RNA editing identification:
2.1 REDItools
python REDItoolDenovo.py -i SRR2079784.sorted.bam -f arabidopsis_mt.fa -o SRR2079784 -c 4 >SRR2079784.nohup &
2.2 REDO
Coconut cp:
perl REDO.pl -v SRR1063404_cp.vcf -v SRR1063407_cp.vcf -v SRR1137438_cp.vcf -v SRR1173229_cp.vcf -v SRR1265939_cp.vcf -v SRR1273070_cp.vcf -v SRR1273180_cp.vcf -v SRR606452_cp.vcf -t cp.tbl -g cp.fsa -o coco_cp &
Arabidopsis mt:
perl REDO.pl -g mt.fa -v SRR2079784_mt.vcf -t mt.tbl -o ath_mt -a 0 -s 0&
Note: because Arabidopsis has a lot of RNA editing sites with short distance (adjacent RNA editing sites) or located in the junction points of transcripts in experiment data, we changed the default parameter -a and -s to 0 to recover these RNA editing sites.
Table 2 The information of RNA editing identification in test RNA-seq data
Species Organelle SRA accession No. Samtools Variant REDItools editing sites Percent REDO editing sites Percent
Coconut cp SRR1063404 210 126 60.00% 96 45.71%
SRR1063407 387 - - 66 17.05%
SRR1137438 197 - - 51 25.89%
SRR1173229 590 - - 129 21.86%
SRR1265939 477 - - 49 10.27%
SRR1273070 1426 375 26.30% 175 12.27%
SRR1273180 229 - - 88 38.43%
SRR606452 247 - - 68 27.53%
Arabidopsis mt SRR2079784 1219 553 45.37% 488 40.03%
3. RNA editing result:
3.1 Figures
Figure 1 - The attributes of RNA editing sites in Arabidopsis mt.
Figure 2 - The attributes of RNA editing sites in coconut SRR1063404.
Figure 3 - The attributes of RNA editing sites in coconut SRR1273070.
Figure 4 - The cluster of RNA editing sites in coconut cp based on 8 RNA-seq data according to the RNA editing proportions.
Figure 5 ¨C The attributes comparison between true positive and false positive RNA editing sites in mt (not produced by this program).
3.2 Tables
Table 3 - ath_mt_SRR2079784_mt.out (partial).
Table 4 - coco_cp.out (partial).
Table 5 - coco_cp.change_matrix.
Table 6 - coco_cp.stat.
Table 7 - coco_cp_DPR.out (partial).
Table 8 - The RNA editing sites identification between REDO and REDItools in three analyzed RNA-seq datasets (without applying the complicated filter model).
Table 9 - The RNA editing sites identification between REDO and REDItools in three analyzed RNA-seq datasets (with applying the complicated filter model).
3.2.1 Table 3 - ath_mt_SRR2079784_mt.out (partial).
#Chr Strand Name Genome_pos Gene_pos AA_pos Phase Ref->Alt RefCodon->AltCodon RefAA->AltAA AltRatio ErrorRatio Depth AverageDepth RefRead AltRead AdjacentAlt LLR Pvalue(fisher) RNAeditGC GeneGC Exon_num Exon_start Exon_end Function
ath - nad2 81279 19 7 1 C->T CGG->TGG R->W 0.18914845516202 0.000623221129939135 1327 1271.3 1076 251 9 1000 1.15601898625219e-81 55.00 39.93 5 81297,80132,333105,330306,328078, 81113,79740,332945,329735,327890, NADH dehydrogenase subunit 2
ath - nad2 81270 28 10 1 C->T CCA->TCA P->S 0.552412645590682 0.000497481358414786 1202 1048.9 538 664 31 1000 1.19926472602608e-256 55.00 39.93 5 81297,80132,333105,330306,328078, 81113,79740,332945,329735,327890, NADH dehydrogenase subunit 2
ath - nad2 81239 59 20 2 C->T TCC->TTC S->F 0.973550356052899 0.00103016776810268 983 930.3 26 957 30 1000 0 55.00 39.93 5 81297,80132,333105,330306,328078, 81113,79740,332945,329735,327890, NADH dehydrogenase subunit 2
ath - nad2 81209 89 30 2 C->T TCC->TTC S->F 0.930120481927711 0.000477897934247639 415 417.6 29 386 1 1000 1.33197151054271e-203 45.00 39.93 5 81297,80132,333105,330306,328078, 81113,79740,332945,329735,327890, NADH dehydrogenase subunit 2
......
3.2.2 Table 4 - coco_cp.out (partial).
#Chr Strand Name Genome_pos Gene_pos AA_pos Sample_num Phase SRR1063404_cp SRR1063407_cp SRR1137438_cp SRR1173229_cp SRR1265939_cp SRR1273070_cp SRR1273180_cp SRR606452_cp Exon_num Exon_start Exon_end Function
cp - rpl23 2143 71 24 2 2 C->T:TCT->TTT:S->F:0.920792079207921:0:202:197.2:16:186:18:1000:4.31666359754032e-97:25.00:37.59 C->T:TCT->TTT:S->F:0.968599033816425:0.000216450216450216:414:451.9:13:401:18:1000:4.2221509942939e-224:25.00:37.59 1 2213, 1932, ribosomal protein L23
cp - trnM-CAU 2399 49 ? 1 ? C->T:?->?:?->?:0.75:0:12:12:3:9:27:28.7754103989239:0.000168259523489029:40.00:45.95 1 2447, 2374, tRNA
cp + ycf2 2886 371 124 1 2 G->A:AGA->AAA:R->K:0.6:0.025:20:19.9:8:12:416:11.3538827969065:2.25475753840605e-05:45.00:37.66 1 2516, 9400, hypothetical chloroplast RF21
cp + ycf2 2997 482 161 2 2 C->T:CCG->CTG:P->L:0.333333333333333:0:18:15:12:6:5:20.8308785164354:0.00953079178885632:50.00:37.66 C->T:CCG->CTG:P->L:0.75:0:12:13:3:9:5:28.7754103989239:0.000168259523489029:50.00:37.66 1 2516, 9400, hypothetical chloroplast RF21
......
3.2.3 Table 5 - coco_cp.change_matrix (number and percent).
#Prefix A->C(percent) A->G(percent) A->T(percent) C->G(percent) C->T(percent) C->A(percent) G->T(percent) G->A(percent) G->C(percent) T->A(percent) T->C(percent) T->G(percent)
SRR1063404_cp 0(0) 0(0) 0(0) 0(0) 96(100) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0) 0(0)
SRR1063407_cp 0(0) 0(0) 0(0) 1(1.52) 59(89.39) 0(0) 0(0) 5(7.58) 0(0) 0(0) 1(1.52) 0(0)
SRR1137438_cp 0(0) 0(0) 0(0) 0(0) 50(98.04) 0(0) 0(0) 0(0) 0(0) 0(0) 1(1.96) 0(0)
SRR1173229_cp 2(1.55) 5(3.88) 0(0) 2(1.55) 103(79.84) 0(0) 0(0) 10(7.75) 0(0) 1(0.78) 3(2.33) 2(1.55)
SRR1265939_cp 0(0) 0(0) 0(0) 0(0) 44(84.62) 2(3.85) 0(0) 2(3.85) 0(0) 0(0) 1(1.92) 0(0)
SRR1273070_cp 5(2.79) 9(5.03) 1(0.56) 1(0.56) 85(47.49) 3(1.68) 3(1.68) 43(24.02) 3(1.68) 2(1.12) 16(8.94) 4(2.23)
SRR1273180_cp 0(0) 0(0) 0(0) 0(0) 87(97.75) 0(0) 0(0) 1(1.12) 0(0) 0(0) 0(0) 0(0)
SRR606452_cp 0(0) 0(0) 1(1.47) 0(0) 65(95.59) 0(0) 0(0) 1(1.47) 0(0) 0(0) 0(0) 1(1.47)
3.2.4 Table 6 - coco_cp.stat #(1:hydrophobic2hydrophobic,2:hydrophilic2hydrophilic,3:hydrophobic2hydrophilic,4:hydrophilic2hydrophobic).
#Prefix Total Other Phase(1,2,3) Silent NonSilent A->C(silent) A->C(non_silent) A->C(1,2,3,4) A->G(silent) A->G(non_silent) A->G(1,2,3,4) A->T(silent) A->T(non_silent) A->T(1,2,3,4) C->G(silent) C->G(non_silent) C->G(1,2,3,4) C->T(silent) C->T(non_silent) C->T(1,2,3,4) C->A(silent) C->A(non_silent) C->A(1,2,3,4) G->T(silent) G->T(non_silent) G->T(1,2,3,4) G->A(silent) G->A(non_silent) G->A(1,2,3,4) G->C(silent) G->C(non_silent) G->C(1,2,3,4) T->A(silent) T->A(non_silent) T->A(1,2,3,4) T->C(silent) T->C(non_silent) T->C(1,2,3,4) T->G(silent) T->G(non_silent) T->G(1,2,3,4)
SRR1063404_cp 96 0 15,73,9 9 87 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 9 87 15,10,1,61 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0
SRR1063407_cp 66 0 14,47,7 7 59 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 1 1,0,0,0 6 53 9,7,1,36 0 0 0,0,0,0 0 0 0,0,0,0 0 5 3,1,1,0 0 0 0,0,0,0 0 0 0,0,0,0 1 0 0,0,0,0 0 0 0,0,0,0
SRR1137438_cp 51 0 10,37,5 4 47 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 3 47 9,6,1,31 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 1 0 0,0,0,0 0 0 0,0,0,0
SRR1173229_cp 129 1 36,82,32 20 108 1 1 1,0,0,0 2 3 2,0,0,1 0 0 0,0,0,0 1 1 1,0,0,0 14 89 18,11,3,57 0 0 0,0,0,0 0 0 0,0,0,0 0 10 4,1,5,0 0 0 0,0,0,0 0 1 1,0,0,0 2 1 1,0,0,0 0 2 0,0,2,0
SRR1265939_cp 52 3 12,37,8 4 45 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 3 41 9,6,0,26 0 2 0,1,1,0 0 0 0,0,0,0 0 2 1,0,1,0 0 0 0,0,0,0 0 0 0,0,0,0 1 0 0,0,0,0 0 0 0,0,0,0
SRR1273070_cp 179 4 64,87,72 49 126 3 2 0,1,0,1 2 7 1,2,0,4 1 0 0,0,0,0 0 1 1,0,0,0 23 62 15,13,4,30 0 3 0,2,1,0 2 1 1,0,0,0 10 33 9,14,10,0 0 3 2,0,0,1 1 1 1,0,0,0 7 9 2,0,6,1 0 4 2,0,2,0
SRR1273180_cp 89 1 15,69,4 5 83 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 5 82 14,10,1,57 0 0 0,0,0,0 0 0 0,0,0,0 0 1 0,0,1,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0
SRR606452_cp 68 0 15,48,8 8 60 0 0 0,0,0,0 0 0 0,0,0,0 1 0 0,0,0,0 0 0 0,0,0,0 6 59 10,6,2,41 0 0 0,0,0,0 0 0 0,0,0,0 0 1 1,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 0 0 0,0,0,0 1 0 0,0,0,0
3.2.5 Table 7 - coco_cp_DPR.out (partial)
Chr Position Proportion1 Alt1 Ref1 Proportion2 Alt2 Ref2 Fold(log2) Pvalue
#SRR1063404_cp vs SRR1063407_cp
cp 136348 0.5 6 6 0 0 0 NA 1
cp 143536 0.936986301369863 342 23 1 24 0 0.09 0.382540126150565
cp 76241 0.571428571428571 40 30 1 7 0 0.81 0.0386037932652878
cp 32743 0.466531440162272 230 263 0.8125 13 3 0.80 0.00902812345271057
......
3.2.6 Table 8 - The RNA editing sites identification between REDO and REDItools in three analyzed RNA-seq datasets (without applying complicated filter model).
Samples Software Type True Total True+ True- False+ False- Precision Sensitivity Specificity
Coconut cp1
(SRR1063404) REDO CTa 75 83 67 33349 16 8 80.72% 89.33% 99.95%
REDItools 92 69 33340 23 6 75.00% 92.00% 99.93%
Common 80 65 33352 15 10 81.25% 86.67% 99.96%
REDO Allb 84 67 33348 17 8 79.76% 89.33% 99.95%
REDItools 98 69 33334 29 6 70.41% 92.00% 99.91%
Common 84 67 33347 17 8 79.76% 89.33% 99.95%
Coconut cp2
(SRR1273070) REDO CT 54 23 33378 31 52 42.59% 30.67% 99.91%
REDItools 57 22 33375 35 53 38.60% 29.33% 99.90%
Common 49 20 33383 29 55 40.82% 26.67% 99.90%
REDO All 112 23 33320 89 52 20.54% 30.67% 99.73%
REDItools 121 22 33311 99 53 18.18% 29.33% 99.70%
Common 98 20 33334 78 55 20.41% 26.67% 99.74%
Arabidopsis mt
(SRR2079784) REDO CT 428c 413 358 29041 55 70 86.68% 83.64% 99.81%
REDItools 430 362 29024 68 66 84.19% 84.58% 99.77%
Common 409 356 29045 53 72 87.04% 83.18% 99.81%
REDO All 430 358 29024 72 70 83.26% 83.64% 99.75%
REDItools 469 362 28985 86 66 80.80% 84.58% 99.70%
Common 426 356 29028 70 72 83.57% 83.18% 99.76%
Note: REDItools was run with ¡°-c 4¡± option (total reads coverage>=4) and the result was filtered by p-value<=0.01. RED was run using ¡°-a 0, -s 0¡± option (minimum alt distance==0 and minimum junction anchor size==0) for mt data and default for cp data without applying the complicated filter model. a: CT means C->T type. b: All means all possible RNA editing types. c: There are 455 C->T RNA editing sites in Arabidopsis by cDNA clones sequencing while 428 C->T RNA editing sites are located in CDS region. We only used these 428 C->T RNA editing sites for mt RNA editing sites test.
3.2.7 Table 9 - The RNA editing sites identification between REDO and REDItools in three analyzed RNA-seq datasets (with applying complicated filter model).
Samples Software Type True Total True+ True- False+ False- Precision Sensitivity Specificity
Coconut cp1
(SRR1063404) REDO CTa 75 83 67 16 8 80.72% 89.33% 33349 99.95%
REDItools 92 69 23 6 75.00% 92.00% 33340 99.93%
Common 80 65 15 10 81.25% 86.67% 33352 99.96%
REDO Allb 83 67 16 8 80.72% 89.33% 33349 99.95%
REDItools 98 69 29 6 70.41% 92.00% 33334 99.91%
Common 80 65 15 10 81.25% 86.67% 33352 99.96%
Coconut cp2
(SRR1273070) REDO CT 44 22 22 53 50.00% 29.33% 33388 99.93%
REDItools 57 22 35 53 38.60% 29.33% 33375 99.90%
Common 39 19 20 56 48.72% 25.33% 29415 99.93%
REDO All 71 22 49 53 30.99% 29.33% 29383 99.83%
REDItools 121 22 99 53 18.18% 29.33% 29333 99.66%
Common 65 19 46 56 29.23% 25.33% 29389 99.84%
Arabidopsis mt
(SRR2079784) REDO CT 428c 406 356 50 72 87.68% 83.18% 29048 99.83%
REDItools 430 362 68 66 84.19% 84.58% 29024 99.77%
Common 402 354 48 74 88.06% 82.71% 29052 99.84%
REDO All 419 356 63 72 84.96% 83.18% 29035 99.78%
REDItools 469 362 86 66 80.80% 84.58% 28985 99.70%
Common 415 354 61 74 85.30% 82.71% 29039 99.79%
Note: REDItools was run with ¡°-c 4¡± option (total reads coverage>=4) and the result was filtered by p-value<=0.01. RED was run using ¡°-a 0, -s 0¡± option (minimum alt distance==0 and minimum junction anchor size==0) for mt data and default for cp data with applying the complicated filter model. a: CT means C->T type. b: All means all possible RNA editing types. c: There are 455 C->T RNA editing sites in Arabidopsis by cDNA clones sequencing while 428 C->T RNA editing sites are located in CDS region. We only used these 428 C->T RNA editing sites for mt RNA editing sites test.
SUBPROGRAM
==========
We also provide a subprogram named fish.pl for processing the REDO result tables easily and quickly. It can be used to find common and unique sites in two samples or remove genome variant in REDO result file by comparing with genome variant file.
It can get unique bait/baits (one or multiple items in column) in bait file at first (such as chromosome, position, strand, and others), then these bait/baits are used to search the corresponding column in fish file in two measures, existing in bait file (de-fault) or not exiting in bait file ("-contrary" option).
CONTACT
=======
If you have any question or suggestion, please contact us.
Wanfei Liu & Qiang Lin
Email: <liuwf@big.ac.cn> & <linq@big.ac.cn>
Download Latest Version
REDO.tar.gz (29.3 MB)
