dna.vim is a plugin to aid visual parsing of sequencing files (e.g., SAM, PAF, anything with a DNA sequence in it). Current features:

1. A/C/G/T/U/N are colored (consistent with IGV colors for ACGT)
2. Using the commands :SAM, :BAM,:GAF, or :PAF in their respective files will tell you the description of the field your cursor is hovering over (e.g., using :SAM in column 2 of a SAM/BAM file will print a message along the lines of "FLAG: 2064 -- supplementary alignment, reverse strand")
3. Operation blocks within CIGAR strings are colored separately from each other
4. Using :Phred will decode the Phred score of the hovered character (e.g., using it on a D will print "D is score 35, 0.0003 probability of error")
5. Sequence names in FASTA/FASTQ files are colored
6. Tags in alignment files are colored

Screenshots

Usage

Copy dna.vim into your ~/.vim/plugin/ directory. The plugin will automatically trigger for files with the following extensions: .sam, .paf, .fasta, .fa, .fastq, .fq, .gaf, .gfa

You can also toggle the highlighting by using the command :DNA. For example, to view .bam files, you can view the file with samtools view file.bam | vim - and then apply :DNA inside vim.

There are also file-specific commands (:SAM, :GAF:, or :PAF) that will print the description of the field the cursor is hovering over. The field descriptions are pulled from each file type's specification. For example, using :SAM in column 1 of a SAM/BAM file will print "QNAME: Query template NAME". Using :SAM in the flag field will also decode the flag (e.g., "FLAG: 2064 -- supplementary alignment, reverse strand").

Similarly, using :Phred will print the meaning of the Phred score of the hovered character. For instance, when the cursor is over a D, :Phred will print "D is score 35, 0.0003 probability of error".

Performance

The line setlocal synmaxcol=9000 limits how far into a line the syntax matching will search. The value 9000 works well for the files I work with, but if the rendering is too slow, you can reduce lag by lowering the value.

More Precise Matching

The default syntax matching is simple—every A/C/G/T will be colored regardless of context (e.g., the G in HG002 would also be colored). You can implement more careful matching with something like the following, which only highlights bases if they are adjacent to another A/C/G/T:

# checks if the character is adjacent to A/C/G/T
call s:DefineHighlight('Adenine', '\([ACGT]\)\@<=A\|A\([ACGT]\)\@=', '#009600', 28)

# default pattern that just checks for A
call s:DefineHighlight('Adenine', 'A', '#009600', 28)

However, the trade-off is that the more precise matching is slower. Since the capital A/C/G/T don't appear too often in sequencing files outside of DNA sequences, the plugin uses the faster simple matching.