Awesome
sam2pairwise
sam2pairwise takes a SAM file and uses the CIGAR and MD tag to reconstruct the pairwise alignment of each read.
Requirements
A C++ compiler, such as G++.
Installation
To compile sam2pairwise, enter the src/ directory and type:
make
sam2pairwise can then be moved to /usr/bin/ or wherever is convenient. To remove the compiled files, enter the src/ directory and type:
make clobber
Usage
Input can be read out of a SAM file:
sam2pairwise < file.sam > file.out
Or from a pipe:
samtools view file.bam | sam2pairwise > file.out
To display the current version and quit, enter:
sam2pairwise --version
Output
sam2pairwise currently supports the following CIGAR elements for reconstructing the alignment:
- M: Alignment match. The MD tag is used to determine if this is a sequence match or mismatch.
- I: Insertion to the reference, indicated by '-' on the reference sequence.
- D: Deletion from the reference, indicated '-' on the read sequence.
- N: Skipped region from the reference, indicated by '.' on the read and 'N' on the reference sequence.
- S: Soft clipping. These characters are represented normally on the read sequence, and as 'N's on the reference sequence.
- H: Hard clipping. These characters are no longer present in the read sequence, so this value is skipped.
- P: Padding. Represented as an * in both the read and reference sequences.
- =: Sequence match. Treated the same as 'M'.
- X: Sequence mismatch. Treated the same as 'M'.
Each line of input has a corresponding four lines of output. The first line serves as the header, and reproduces the first nine mandatory fields from the SAM entry. The second line is the read itself, the third indicates sequence matches, and the fourth represents the reference sequence.
So, an input like this:
CRATOS:145:D1UH5ACXX:2:1308:6211:53528 153 Zv9_scaffold3453 49562 26 1S15M1D65M2I18M = 49562 0 GCCTGAGAACAAGTGAGAAAGAAACTCATTCCTGTCTTTCAATGAGTGCTTTTGTGCATTTAGGAGAACTAGGCAGCACACATTTAGGGCTGAAAGATGNA (CCDCCCCAEECCDFFFFFFFHECHFHJIHDJIIIIJJJJJJJJJJJJIHGIFJJIGJJJIIIIIIJIJIJIGIIHGFCCJJJJJIJJHGHHHFFFDA1#C PG:Z:novoalign AS:i:206 UQ:i:206 NM:i:7 MD:Z:15^T35A30C7C6G1
Would produce an output like this:
CRATOS:145:D1UH5ACXX:2:1308:6211:53528 153 Zv9_scaffold3453 49562 26 1S15M1D65M2I18M = 49562 0
GCCTGAGAACAAGTGA-GAAAGAAACTCATTCCTGTCTTTCAATGAGTGCTTTTGTGCATTTAGGAGAACTAGGCAGCACACATTTAGGGCTGAAAGATGNA
||||||||||||||| ||||||||||||||||||||||||||||||||||| ||||||||||||||||||||||||||||| | ||||||| |||||| |
NCCTGAGAACAAGTGATGAAAGAAACTCATTCCTGTCTTTCAATGAGTGCTTATGTGCATTTAGGAGAACTAGGCAGCACAC--TCAGGGCTGCAAGATGGA
In the event that the MD tag is not detected & the CIGAR contains characters that require the MD tag to resolve, or if an unsupported character is detected in the CIGAR, the program maintains the four-line periodicity of the output by printing the header output, an error message, and two blank lines.
Citation
If you use sam2pairwise in your work, please use the following DOI to cite it:
Contact
Matthew C. LaFave, Ph.D. Developmental Genomics Section, Translational and Functional Genomics Branch NHGRI, NIH
Email: matthew.lafave [at sign] nih.gov