ve-bamreader

0.2.5 • Public • Published

Note: Forked from shinout bamreader (all credit goes to shinout)

BAMLineReader

BAMLineReader is a reader to parse .bam files. Uses samtools If exists, otherwise uses a native parser.

installation

$ npm install ve-bamreader 
# May require CC and CXX environmental variables on certain distributions pointing to `Nan` compatible GCC version.  
# 4.9.0 has worked in the past 

usage

var bamreader = require("ve-bamreader");
 
bamreader("/path/to/bamfile.bam").then((jsonResult) => {
  //do something with the resulting json :)
})

streaming usage (you probably don't need this)

var BAMLineReader = require("bamreader").BAMLineReader;
var reader = BAMLineReader.create("/path/to/bamfile.bam");
reader.on("bam", function(bam) {
  // bam: object. see "bam object" section
  console.log(bam.seq, bam.qual);
  console.log(bam.pair); // pair bam object of the bam. To do this, indexing is needed.
});
 
reader.on("end", function() {
  console.log("all bam alignments have been read.");
});

bam object

  • pair : information of the mate pair. see "bam.pair"
  • discordant : discordant or not. see "bam.discordant"
  • qname : name of the read
  • flag : bitwise flag
  • rname : reference sequence rname
  • pos : 1-based leftmost mapping position
  • mapq : mapping quality
  • cigar : CIGAR string
  • rnext : reference sequence name of the primary alignment of the next read
  • next_rname : the actual reference name of the next read.
  • pnext : position of the primary alignment of the next read
  • tlen : template length
  • seq : segment jsequence
  • qual : ASCII of base quality plus 33
  • start : 0-based leftmost mapping position
  • flags : information of flags
  • tagstr : optional fields as string
  • tags : optional fields as structured data described in "bamdata.tags" section
  • multiple : information from flag: template have multiple segment or not.
  • unmapped : information from flag: unmapped or not.
  • allmatches : information from flag: each segment properly aligned or not
  • next_unmapped : information from flag: next segment is unmapped or not.
  • reversed : information from flag: reversely mapped or not.
  • next_reversed : information from flag: next segment is reversely mapped or not.
  • first : information from flag: first segment or not.
  • last : information from flag: last segment or not.
  • secondary : information from flag: secondary alignment or not.
  • lowquality : information from flag: not passing quality control or not.
  • duplicate : information from flag: PCR or optic duplicate or not.
  • supplementary : information from flag: supplementary alignment or not.
  • clipped : clipped or not.
  • soft_clipped : soft-clipped or not.
  • hard_clipped : hard-clipped or not.
  • left_break : the left position of the breakpoint if clipped.
  • right_break : the right position of the breakpoint if clipped.
  • match_len : length of the matched portion of the read.
  • fully_matched : if the whole base of the read is matched or not
  • mismatch : the number of mismatch bases
  • unique : if the read is mapped uniquely or not
  • indel : the mapping result contains insertion/deletion or not.
  • different_ref : if both mapped and rname is not rnext or not
  • same_strand : if both mapped and mapped to the same strand or not
  • has_n : the read contains "N" or not
  • sam : SAM formatted string of the bam

bam.pair

bamreader can fetch one mate pair of the read using index.

$ bamreader -c <bamfile>     # creates an index of the bamfile. 

then you can access bam.pair

bam.pair is also a bam object.

bam.discordant

(...)

bam.tags

  • (key): tag name (two string character)
  • (value): type: data type, value: data

if data type is "B", value[0] is subtype, value.slice(1) is the array of data.

Dependencies (2)

Dev Dependencies (1)

Package Sidebar

Install

npm i ve-bamreader

Weekly Downloads

3

Version

0.2.5

License

none

Unpacked Size

331 kB

Total Files

30

Last publish

Collaborators

  • teselagen-admin
  • tnrich