Splice-Break2: Mitochondrial DNA Deletion Pipeline with Filtering, Annotation, and Batch Processing Capabilities

Splice-Break2 is updated to include bbmap processing, MapSplice 2 alignment, filtering steps, and MITOMAP annotations not presented in the previous version. Splice-Break2 is also capable of batch processing a large number of fastq files when presented in the same directory, can be used with both single-end and paired-end datasets, and includes an alternate reference sequence option to handle a uncommon (but possible) segmentation fault error that we’ve observed in a small number of NGS flowcell runs.

For a detailed description of methods, see:

• Hjelm et al. “Splice-Break: exploiting an RNA-seq splice junction algorithm to discover mitochondrial DNA deletion breakpoints and analyses of psychiatric disorders”. https://doi.org/10.1093/nar/gkz164
• Xu et al. “Splice-Break2: Mitochondrial DNA Deletion Pipeline with Filtering, Annotation, and Batch Processing Capabilities” [link TBD]

Dependencies

Provided within Splice-Break2 Directory

bbmap
MapSplice 2 (v2.2.0+)
Samtools (v1.8+)

Not Provided

Python (v2.7.5+)
Java (v1.8.0+)

Installation

• Download Splice-Break2-v3.0.1_PAIRED-END.tar.gz or Splice-Break2-v3.0.1_SINGLE-END.tar.gz
• Extract Directory: tar -xvf Splice-Break2-v3.0.1_PAIRED-END.tar.gz or tar -xvf Splice-Break2-v3.0.1_SINGLE-END.tar.gz
• Display <SB_PATH>: readlink -f Splice-Break2-v3.0.1_PAIRED-END or readlink -f Splice-Break2-v3.0.1_SINGLE-END

Usage

Command Line

Paired-End

./Splice-Break2_paired-end.sh <inputDir> <outputDir> <logDir> <SB_Path> [options]

Single-End

./Splice-Break2_single-end.sh <inputDir> <outputDir> <logDir> <SB_Path> [options]

Input Files [FASTQ(s)]

Paired-End

• Files must end with a read identifier (.R1/.R2; .r1/.r2; .READ1/.READ2; .read1/.read2) and extension (.fq, .fastq, .txt). A period must be placed before the read identifier and extension. Examples are shown below (all possible combinations of read identifiers and extensions are acceptable).
  • <sample>.read1.fq and <sample>.read2.fq
  • <sample>.R1.fastq and <sample>.R2.fastq
  • <sample>.READ1.txt and <sample>.READ2.txt

Single-End

• Files must end with an extension (.fq, .fastq, .txt). A period must be placed before the extension. Examples are shown below.
  • <sample>.fq
  • <sample>.fastq
  • <sample>.txt

Command Line (Required Directories)

1. <inputDir>: path to directory containing FASTQ file(s)
2. <outputDir>: path to directory where output files will be written
3. <logDir>: path to directory where log files will be written
4. <SB_Path>: path to Splice-Break2 parent directory (/PATH/Splice-Break2_v3.0.0_PAIRED-END or /PATH/Splice-Break2_v3.0.0_SINGLE-END)

Command Line (Options)

• --align=[yes/no]:
  • --align=yes is DEFAULT.
  • --align=yes will align fastq files with MapSplice 2 and proceed with annotation and filtering steps of the Splice-Break2 pipeline.
  • --align=no will only do pre-alignment steps (bbmap; PCR replicate and adapter removal) but process will stop before MapSplice 2 alignment.
• --ref=[rCRS/Nsub]:
  • --ref=rCRS is DEFAULT.
  • --ref=rCRS will use the revised Cambridge Reference Sequence (rCRS; NC_012920.1) of the human mitochondrial genome for alignment.
  • --ref=Nsub will use a modified version of the rCRS sequence, where the first and last 100 bases have been replaced with N’s so that reads cannot align to these regions. We typically only use the NSub option if initial attempts to align with rCRS fail, and this option is suggested in the log file when that occurs.
• --fastq_keep=[yes/no]:
  • --fastq_keep=no is DEFAULT.
  • --fastq_keep=no will remove the replicate FASTQ files in the output directory that have the PCR replicates and adapter removed (e.g., <sample>.deduped.fq and <sample>.deduped.deAdpat.fq).
  • --fastq_keep=yes will retain these files for the user. The number (and %) of reads containing PCR replicates and adapter sequence that were removed from the original FASTQ files are recorded in the log file <logDir>, regardless if these files are kept or not.
• --skip_preAlign=[yes/no]:
  • --skip_preAlign=no is DEFAULT.
  • --skip_preAlign=no will use bbmap to remove PCR replicates and adapter sequences from FASTQ files prior to alignment.
  • --skip_preAlign=yes will skip these steps (not suggested unless steps were performed with alternate program).

Note: If the user wishes to change any of the command options from the DEFAULT settings (above), all four command options must be specified in ORDER (--align, --ref, --fastq_keep, --skip_preAlign).

Description of Files in <logDir> and <outputDir>

<logDir>

• <sample>.log
  • The <sample>.log file contains the input and output read numbers (and %’s) after PCR replicate and adapter removal from bbmap processes, the MapSplice 2 log file information, and the Splice-Break2 processes, filtering and annotation steps.

<outputDir>

• <sample>-READ*.deduped.fq
  • The <sample>-READ*.deduped.fq files are the FASTQs after PCR replicate removal (bbmap). File(s) will only be present if --fastq_keep=yes in command line.
• <sample>-READ*.deduped.deAdapt.fq
  • The <sample>-READ*.deduped.deAdapt.fq files are the FASTQs after TruSeq adapter removal (bbmap). File(s) will only be present if --fastq_keep=yes in command line.
• <sample>_mapsplice-timestamp [directory with results; see below for contents]

Description of Files in <outputDir>/<sample>_mapsplice-timestamp

Large mtDNA Deletion Files

• <sample> _LargeMTDeletions_LR-PCR_STRINGENT_pos357-15925.txt
  • <sample> _LargeMTDeletions_LR-PCR_STRINGENT_pos357-15925.txt is the file containing the large mtDNA deletion calls, after removal of deletions that had 5’ or 3’ breakpoints within 500bp of the described primers. We suggest using this file for downstream analysis when a long-range PCR was performed for mitochondrial enrichment. Matches filtering described in the Splice-Break paper (https://doi.org/10.1093/nar/gkz164).
• <sample>_LargeMTDeletions_LR-PCR_conservative_pos157-16125.txt
  • <sample>_LargeMTDeletions_LR-PCR_conservative_pos157-16125.txt is the file containing the large mtDNA deletion calls, after removal of deletions that had 5’ or 3’ breakpoints within 250bp of the described primers. Use with caution as we have observed primer read peaks and false-positive calls that extend past this position.
• <sample>_LargeMTDeletions_WGS-only_NoPositionFilter.txt
  • <sample>_LargeMTDeletions_WGS-only_NoPositionFilter.txt is the file containing the large mtDNA deletion calls, when no filtering is done with respect to primer position. This file should only be used for whole genome sequencing (WGS) data when no long-range PCR was performed for mitochondrial enrichment.
• <sample>_LargeMTDeletions_DNAorRNA_Top30_NARpub.txt
  • <sample>_LargeMTDeletions_DNAorRNA_Top30_NARpub.txt is the file containing only the “Top 30 deletions” described in the Splice-Break paper (https://doi.org/10.1093/nar/gkz164). This file can be used to examine a small list of high-frequency mtDNA deletions that have been experimentally validated, and can be used for both DNA and RNA datasets. See [link TBD] for additional information on applicable RNA-Seq data.
• The Column Description of LargeMTDeletions files:

  • Sample_ID	MapSplice_Breakpoint	5'_Break	3'_Break	Deletion_Size_bp	Deletion_Reads	Benchmark_Coverage	Deletion_Read_%	Annotation	Left_Overhang	Right_Overhang	IMPACT_MITOMAP
    the ID of Sample	...	...	...	...	...	...	...	...	...	...	...

Coverage and Benchmark Files

• <sample>_Coverage.txt
  • The <sample>_Coverage.txt file is a table with coverage (depth) at each base of the rCRS (NC_012920.1) sequence. Primer positions are also annotated, but are only accurate if primers described in the Splice-Break paper were used (5′ -CCGCACAAGAGTGCTACTCTCCTC-3′ and 5′ -GATATTGATTTCACGGAGGATGGTG-3′). This file can be used to create a coverage plot. We suggest plotting Primer_Position on the X-axis and Coverage on the Y-axis, so that plot matches the PCR-amplified linear molecules (if described primers were used).
• Benchmark.txt
  • Benchmark.txt has the average coverage (depth) across the benchmark positions described in the Splice-Break paper (https://doi.org/10.1093/nar/gkz164). This value is used for normalization (% calculations) of the large mtDNA deletions, and is also provided within the ‘<sample>_LargeMTDeletions_’ files. We require this number to be > or = to 5,000x for most studies.

MapSplice 2 Output Files

• <sample>_alignments.bam
  • The <sample>_alignments.bam file is the sorted BAM file after MapSplice 2 alignment.
• stats.txt
  • The stats.txt file has the read alignment statistics from MapSplice 2.
• logs [directory]
  • The logs [directory] contains the log files from MapSplice 2. Important log information for alignment steps is also provided in the <sample>.log file located in <logDir>.
• junctions.txt
  • The junctions.txt file is the initial output of “spliced” reads from MapSplice 2, which is used for Splice-Break2 processing. User should not use this file as will contain false-positives; provided as reference.

Updates

Version 3.0.1 -- scripting of LargeMTDeletions_DNAorRNA_Top30_NARpub.txt updated

Versioning

Semantic Versioning 3.0.1

Contact

Brooke Hjelm
bhjelm@usc.edu

Lili Xu
lilixu@usc.edu

Michelle Webb
michelgw@usc.edu