/// <summary>
/// Aligns reads and outputs alignment map and chimeric alignments.
/// Note: fastqs must have \n line endings, not \r\n.
/// </summary>
/// <param name="spritzDirectory"></param>
/// <param name="threads"></param>
/// <param name="genomeDir"></param>
/// <param name="fastqFiles"></param>
/// <param name="outprefix"></param>
/// <param name="strandSpecific"></param>
/// <param name="genomeLoad"></param>
/// <param name="outSamType"></param>
/// <returns></returns>
public static List <string> BasicAlignReadCommands(string spritzDirectory, int threads, string genomeDir, string[] fastqFiles, string outprefix, bool strandSpecific = true, STARGenomeLoadOption genomeLoad = STARGenomeLoadOption.NoSharedMemory, string outSamType = "BAM Unsorted")
{
string reads_in = "\"" + string.Join("\" \"", fastqFiles.Select(f => WrapperUtility.ConvertWindowsPath(f))) + "\"";
string read_command = fastqFiles.Any(f => Path.GetExtension(f) == ".gz") ?
" --readFilesCommand zcat -c" :
fastqFiles.Any(f => Path.GetExtension(f) == ".bz2") ?
" --readFilesCommand bzip2 -c" :
"";
string arguments =
" --genomeLoad " + genomeLoad.ToString() +
" --runThreadN " + threads.ToString() +
" --genomeDir \"" + WrapperUtility.ConvertWindowsPath(genomeDir) + "\"" +
" --readFilesIn " + reads_in +
" --outSAMtype " + outSamType +
" --limitBAMsortRAM " + Process.GetCurrentProcess().VirtualMemorySize64.ToString() +
" --outSAMstrandField intronMotif" + // adds XS tag to all alignments that contain a splice junction
" --outFilterIntronMotifs RemoveNoncanonical" + // for cufflinks
" --outFileNamePrefix " + WrapperUtility.ConvertWindowsPath(outprefix) +
read_command;
string fileToCheck = WrapperUtility.ConvertWindowsPath(outprefix + (outSamType.Contains("Sorted") ? SortedBamFileSuffix : outSamType.Contains("Unsorted") ? BamFileSuffix : SpliceJunctionFileSuffix));
return(new List <string>
{
WrapperUtility.ChangeToToolsDirectoryCommand(spritzDirectory),
"if [[ ( ! -f " + WrapperUtility.ConvertWindowsPath(fileToCheck) + " || ! -s " + WrapperUtility.ConvertWindowsPath(fileToCheck) + " ) ]]; then STAR" + arguments + "; fi",
File.Exists(outprefix + BamFileSuffix) && genomeLoad == STARGenomeLoadOption.LoadAndRemove ? "STAR --genomeLoad " + STARGenomeLoadOption.Remove.ToString() : ""
});
}
/// <summary>
/// Aligns reads and outputs alignment map and chimeric alignments. Duplicate reads are removed (deduped) from the alignment map, a step that's recommended for variant calling.
/// Note: fastqs must have \n line endings, not \r\n.
/// </summary>
/// <param name="spritzDirectory"></param>
/// <param name="threads"></param>
/// <param name="genomeDir"></param>
/// <param name="fastqFiles"></param>
/// <param name="outprefix"></param>
/// <param name="strandSpecific"></param>
/// <param name="genomeLoad"></param>
/// <returns></returns>
public static List <string> AlignRNASeqReadsForVariantCalling(string spritzDirectory, int threads, string genomeDir, string[] fastqFiles,
string outprefix, bool overwriteStarAlignment, bool strandSpecific = true, STARGenomeLoadOption genomeLoad = STARGenomeLoadOption.NoSharedMemory)
{
string reads_in = string.Join(" ", fastqFiles.Select(f => WrapperUtility.ConvertWindowsPath(f)));
string read_command = fastqFiles.Any(f => Path.GetExtension(f) == ".gz") ?
" --readFilesCommand zcat -c" :
fastqFiles.Any(f => Path.GetExtension(f) == ".bz2") ?
" --readFilesCommand bzip2 -c" :
"";
string alignmentArguments =
" --genomeLoad " + genomeLoad.ToString() +
" --runMode alignReads" +
" --runThreadN " + threads.ToString() +
" --genomeDir " + WrapperUtility.ConvertWindowsPath(genomeDir) +
" --readFilesIn " + reads_in +
" --outSAMtype BAM SortedByCoordinate" +
" --outBAMcompression 10" +
" --limitBAMsortRAM " + Process.GetCurrentProcess().VirtualMemorySize64.ToString() +
" --outFileNamePrefix " + WrapperUtility.ConvertWindowsPath(outprefix) +
// chimeric junction settings
//" --chimSegmentMin 12" +
//" --chimJunctionOverhangMin 12" +
//" --alignSJDBoverhangMin 10" +
//" --alignMatesGapMax 100000" +
//" --alignIntronMax 100000" +
//" --chimSegmentReadGapMax 3" +
//" --alignSJstitchMismatchNmax 5 -1 5 5" +
// stringtie parameters
" --outSAMstrandField intronMotif" + // adds XS tag to all alignments that contain a splice junction
" --outFilterIntronMotifs RemoveNoncanonical" + // for cufflinks
// gatk parameters
" --outSAMattrRGline ID:1 PU:platform PL:illumina SM:sample LB:library" + // this could shorten the time for samples that aren't multiplexed in preprocessing for GATK
" --outSAMmapqUnique 60" + // this is used to ensure compatibility with GATK without having to use the GATK hacks
read_command; // note in the future, two sets of reads can be comma separated here, and the RGline can also be comma separated to distinguish them later
return(new List <string>
{
WrapperUtility.ChangeToToolsDirectoryCommand(spritzDirectory),
overwriteStarAlignment ? "" :
"if [[ ( ! -f " + WrapperUtility.ConvertWindowsPath(outprefix + SortedBamFileSuffix) + " || ! -s " + WrapperUtility.ConvertWindowsPath(outprefix + SortedBamFileSuffix) + " ) ]]; then",
" STAR" + alignmentArguments,
overwriteStarAlignment ? "" : "fi",
SamtoolsWrapper.IndexBamCommand(WrapperUtility.ConvertWindowsPath(outprefix + SortedBamFileSuffix)),
overwriteStarAlignment ? "" : "if [[ ( ! -f " + WrapperUtility.ConvertWindowsPath(outprefix + DedupedBamFileSuffix) + " || ! -s " + WrapperUtility.ConvertWindowsPath(outprefix + DedupedBamFileSuffix) + " ) ]]; then",
" " + StarDedupCommand(threads, outprefix + SortedBamFileSuffix, outprefix + Path.GetFileNameWithoutExtension(SortedBamFileSuffix)),
overwriteStarAlignment ? "" : "fi",
SamtoolsWrapper.IndexBamCommand(WrapperUtility.ConvertWindowsPath(outprefix + DedupedBamFileSuffix)),
File.Exists(outprefix + BamFileSuffix) && File.Exists(outprefix + DedupedBamFileSuffix) && genomeLoad == STARGenomeLoadOption.LoadAndRemove ?
"STAR --genomeLoad " + STARGenomeLoadOption.Remove.ToString() :
"",
});
}