Data format for RNA-seq data

Calculate gene counts (Tutorial). Input file should include header line and more than 3 fields (1~2). Fill all the blanks with "NA" (blank is not allowed). Save the table as Microsoft Excel (.xls, .xlsx) or tab-delimited text file.
You can get a sample file by clicking on this picture.

Tutorial1: Calculation of Gene counts using Subread

1. Install R.
2. Copy and paste following commands to R console.
2.1. Installation of Rsubread 

x <- library(character.only=T)
if(length(grep("^Rsubread", x$results[,1], value=T))==0){ #Checking installation of Rsubread
source("http://bioconductor.org/biocLite.R")
biocLite("Rsubread")
}
2.2. Build index for a reference genome 

if(length(grep("^RCurl", x$results[,1], value=T))==0){ #Checking installation of RCurl
source("http://bioconductor.org/biocLite.R")
biocLite("RCurl")
}
library(RCurl)
url <- "ftp://ussd-ftp.illumina.com/Arabidopsis_thaliana/Ensembl/TAIR10/Arabidopsis_thaliana_Ensembl_TAIR10.tar.gz"
userpwd <- "igenome:G3nom3s4u"
bin = getBinaryURL(url, userpwd = userpwd,
             ftp.use.epsv = FALSE) 
writeBin(bin, "Arabidopsis_thaliana_Ensembl_TAIR10.tar.gz") 
untar("Arabidopsis_thaliana_Ensembl_TAIR10.tar.gz")

library(Rsubread)
buildindex(basename="arabidopsis_index", reference="./Arabidopsis_thaliana/Ensembl/TAIR10/Sequence/WholeGenomeFasta/genome.fa")
2.3. Download RNA-seq data (optional, replace "SRP051531" with the data of your interest)
If you have your own read data, DO NOT USE this command. Command "getwd()" in R, copy your fastq or fastq.gz files to a directory returned by the command. 

##SRAdb install
if(length(grep("^SRAdb", x$results[,1], value=T))==0){ #Checking installation of SRAdb
source("http://bioconductor.org/biocLite.R")
biocLite("SRAdb")
}
##Download fastq files (in SRA project SRP003951 for example)
library(SRAdb)
sqlfile <- "SRAmetadb.sqlite"
if(!file.exists("SRAmetadb.sqlite")){ sqlfile <- getSRAdbFile() }
sra_con <- dbConnect(SQLite(),sqlfile)
rs = getFASTQinfo ( c("SRP003951"), srcType = 'ftp', sra_con = sra_con ) # Replace this line
getSRAfile( rs$run, sra_con, fileType = 'fastq')
2.4. align Read mapping for RNA-seq data via seed-and-vote 

library(Rsubread)
#List fastq files
fastq.files <- list.files(".", pattern="fastq")
paired.fastq.files.1 <- fastq.files[grep("_1.",fastq.files)]
paired.fastq.files.2 <- fastq.files[grep("_2.",fastq.files)]
single.fastq.files <- fastq.files[grep("_1.|_2.",fastq.files, invert=T)]
##Mapping of single-end reads
for(fastq.file in single.fastq.files){
output.file <- sprintf("%s_subread_results.bam", strsplit(fastq.file, ".fastq")[[1]][1])
align(index="arabidopsis_index", readfile1=fastq.file, output_file=output.file)
}
##Mapping of paired-end reads
if(length(paired.fastq.files.1)==length(paired.fastq.files.2)){
num.pairs <- length(paired.fastq.files.1)
for(j in 1:num.pairs){
output.file <- sprintf("%s_subread_results.bam", strsplit(paired.fastq.files.1[j], "_1.")[[1]][1])
align(index="arabidopsis_index", readfile1=paired.fastq.files.1[j], readfile2=paired.fastq.files.2[j], output_file=output.file, nthreads=3)
}
}else{
cat("Check that all fastq files are paired\n")
}
2.5. featureCounts: a general-purpose read summarization function
Read counts will be recorded in "counts.txt". Command "getwd()" in R to get where the file is. Input this file to AtCAST.

##single-end reads
input.files <- NULL
if(length(single.fastq.files) > 0){
for(i in 1:length(single.fastq.files)){
	input.files[i] <- sprintf("%s_subread_results.bam", strsplit(single.fastq.files[i], ".fastq")[[1]][1])
}
list <- featureCounts(files=input.files, annot.ext="./Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Genes/genes.gtf", isGTFAnnotationFile=TRUE)
}
##paied-end reads
if(length(paired.fastq.files.1) > 0 && length(paired.fastq.files.1)==length(paired.fastq.files.2)){
	num.pairs <- length(paired.fastq.files.1)
	for(j in 1:num.pairs){
		input.files[j] <- sprintf("%s_subread_results.bam", strsplit(paired.fastq.files.1[j], "_1.")[[1]][1])
	}
	list <- featureCounts(files=input.files, annot.ext="./Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Genes/genes.gtf", isPairedEnd=TRUE, isGTFAnnotationFile=TRUE)
}else{
	cat("Check that all fastq files are paired\n")
}
write.table(list, file = "counts.txt", sep = "\t")

Tutorial2: Calculation of Gene counts using cufflinks

1. Install boost, sra_sdk, bowtie2, sam-tools, tophat2 and cufflinks following instructions of tophat2 and cufflinks.
2. Download Arabidopsis ENSEMBL TAIR10 genome and annotation files (compressed in a file) from http://tophat.cbcb.umd.edu/igenomes.shtml, and decompress in a working folder.
3. Prepare your RNA-seq data. (Example we use: NGS seq files (SRR520237.sra, SRR520238.sra, SRR520239.sra, SRR520246.sra, SRR520247.sra, SRR520248.sra) were downloaded from SRA(NCBI) in the same folder and renamed (wt1_SRR520237.sra, wt2_SRR520238.sra, wt3_SRR520239.sra, pifq1_SRR520246.sra, pifq2_SRR520247.sra, pifq3_SRR520248.sra)).
4. Map sequences to Arabidopsis genome. (Example: Execute following commands. Length of reads in the example files were 50 bp.) 
fastq-dump wt1_SRR520237.sra && # This line is not necessary if you use fastq file of your own.
tophat -r 50 -G ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Archives/archive-2011-08-30-20-11-05/Genes/genes.gtf -o tophat_wt_rep1 ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Sequence/Bowtie2Index/genome wt1_SRR520237.fastq && 
fastq-dump pifq1_SRR520246.sra && # This line is not necessary if you use fastq file of your own.
tophat -r 50 -G ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Archives/archive-2011-08-30-20-11-05/Genes/genes.gtf -o tophat_pifq_rep1 ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Sequence/Bowtie2Index/genome pifq1_SRR520246.fastq &

fastq-dump wt2_SRR520238.sra && # This line is not necessary if you use fastq file of your own.
tophat -r 50 -G ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Archives/archive-2011-08-30-20-11-05/Genes/genes.gtf -o tophat_wt_rep2 ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Sequence/Bowtie2Index/genome wt2_SRR520238.fastq && 
fastq-dump pifq2_SRR520247.sra && # This line is not necessary if you use fastq file of your own.
tophat -r 50 -G ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Archives/archive-2011-08-30-20-11-05/Genes/genes.gtf -o tophat_pifq_rep2 ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Sequence/Bowtie2Index/genome pifq2_SRR520247.fastq &

fastq-dump wt3_SRR520239.sra && # This line is not necessary if you use fastq file of your own.
tophat -r 50 -G ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Archives/archive-2011-08-30-20-11-05/Genes/genes.gtf -o tophat_wt_rep3 ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Sequence/Bowtie2Index/genome wt3_SRR520239.fastq && 
fastq-dump pifq3_SRR520248.sra && # This line is not necessary if you use fastq file of your own.
tophat -r 50 -G ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Archives/archive-2011-08-30-20-11-05/Genes/genes.gtf -o tophat_pifq_rep3 ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Sequence/Bowtie2Index/genome pifq3_SRR520248.fastq &
5. Calculate counts using cufflinks using folloing commands. A file generated as "genes.count_tracking" can be applied to AtCAST3. 
cuffdiff ./Arabidopsis_thaliana_Ensembl_TAIR10/Arabidopsis_thaliana/Ensembl/TAIR10/Annotation/Genes/genes.gtf tophat_wt_rep1/accepted_hits.bam tophat_wt_rep2/accepted_hits.bam tophat_wt_rep3/accepted_hits.bam tophat_pifq_rep1/accepted_hits.bam tophat_pifq_rep2/accepted_hits.bam tophat_pifq_rep3/accepted_hits.bam &&
close