bio.io

bio.io

Collection of JavaScript module to parse biological formats.

Gotchas

• Remember that you need CORS to download files from other servers on the web

FASTA

var Fasta = require('biojs-io-fasta');

read(url)

Parses an url an calls your parse method with the returned body.

Fasta.read("https://raw.githubusercontent.com/biojs-io/biojs-io-fasta/master/test/foo.fasta", function(err, model) {
	// model is the parsed url
});

If callback is undefined, read returns a promise.

var p = Fasta.read("https://raw.githubusercontent.com/biojs-io/biojs-io-fasta/master/test/foo.fasta");
// ...
p.then(function(model) {
	// model is the parsed url
}, function(err){
	console.error("err happened during downloading", err);
});

parse(str)

var seqs = Fasta.parse(str);

export(model)

var text = Fasta.export(seqs);

extend(customParser)

Thanks to @sillitoe you can provide a custom parser if your FASTA header is formatted "in weird and wonderful ways".

var customGetMeta = function(header) {
	return {
      id: "id",
      name: "name",
      // optional
      details: {
		foo: "bar"
      },
      // optional
      ids: {
		customId: "bar"
      }
    };
}
var altFasta = Fasta.extend(customGetMeta);

Take a look at tests for a better example of such a parser;

Sequence object

{
  seq: "ABC",
  name: "awesome seq",
  id: "unique id"
}

Clustal

read(url)

Parses an url an calls your parse method with the returned body.

Clustal.read("https://raw.githubusercontent.com/greenify/biojs-io-clustal/blob/master/test/p53.clustalo.clustal", function(err, model) {
	// model is the parsed url
});

If callback is undefined, read returns a promise.

Clustal.read("https://raw.githubusercontent.com/greenify/biojs-io-clustal/blob/master/test/p53.clustalo.clustal").then(function(model) {
	// model is the parsed url
}, function(err){
	console.error("err happened during downloading", err);
});

function is your async callback.

parse(str)

var seqs = Clustal.parse(str);

Sequence object

this.seq
this.name
this.id

Seq.tools

st.getMeta(label)

st.getMeta("sp|abc|def") // {name: "def", ids: {sp: "abc"}}

st.contains(text,search)

st.contains("ABC", 2) // false
st.contains("ABC","C") // true

st.splitNChars(text,n)

st.splitNChars("ABC", 2) // ["AB", "C"]

GFF (General feature format)

A GFF (general feature format) parser

Official Spec

<seqname> <source> <feature> <start> <end> <score> <strand> <frame> [attributes] [comments]

Short description about the formats.

Supported formats

• GFF 3
• Jalview feature format

Getting Started

Install the module with: npm install biojs-io-gff

var gff = require('biojs-io-gff');

Documentation

.read(file, cb)

Callback with parseSeqs or Promise

var p = gff.read("https://cdn.rawgit.com/greenify/biojs-io-gff/master/test/import.gff3");
// ..
p.then(function(seqs){
  // handle the model
}, function(err){
	console.warn(err);
});

.parseSeqs(str) (alias: parse)

Parameter: GFF file (as string) Type: String Example: SEQ1 EMBL atg 103 105 . + 0

Returns a dictionary of all sequences. Each sequences is an array of its features.

gff.parseSeqs('SEQ1  EMBL  atg  103  105  .  +  0');

Result

{ "seqs":
  { "SEQ1":
		[ { seqname: 'SEQ1',
		    source: 'EMBL',
		    feature: 'atg',
		    start: 103,
	    	end: 105,
	    	strand: '+',
	    	frame: 0,
	    	attributes: {} } ]
  },
  "config": {
	type: "gff3"
  }
}

.parseLines(str)

Parameter: GFF file Type: String Example: SEQ1 EMBL atg 103 105 . + 0

The 'parse' method converts a GFF into its JSON representation.

How to use this method

gff.parseLines('SEQ1  EMBL  atg  103  105  .  +  0');

Result

{ "features":
	[{ seqname: 'SEQ1',
    	source: 'EMBL',
    	feature: 'atg',
    	start: 103,
    	end: 105,
    	strand: '+',
    	frame: 0,
    	attributes: {} } ],
  "config": {
	type: "gff3"
  }
}

.exportLines(lines)

Return the textual GFF representation for the given lines

.exportSeqs(seqs) (alias: export)

Return the textual GFF representation for the given seqs

.parseLine(line)

Parameter: GFF line Type: String Example: SEQ1 EMBL atg 103 105 . + 0

The 'parseLine' method converts a GFF line into its JSON representation.

gff.parseLine('SEQ1  EMBL  atg  103  105  .  +  0');

Gotchas

• undefined properties (dots) are removed (checking for undefined is native)

Parser API

It expects that you provide at least a method parse (see below for more details).

Provided methods

read(url)

Parses an url an calls your parse method with the returned body.

parser.read("http://your-url", function(err, model) {
	// model is the parsed url
});

If callback is undefined, read returns a promise.

parser.read("http://your-url").then(function(model) {
	// model is the parsed url
}, function(err){
	console.error("err happened during downloading", err);
});

Expected methods

Your parser should have the following methods:

• parse: Takes in an entire file as string and returns the JSON representation

Optional:

• write: Takes the JSON representation of a file and writes it in the custom format

If the file is line-by-line, one should create a new instance of the parser:

• parseLine: parses another line
• result: returns the current, resulting object of the parsing process.

How to extend

With functions

var parser = function(){
  this.parse = function(data){
      return data;
  };
  Parser.mixin(this);
};

With objects

var throughParserAlt = {
  parse: function(data) {
    return data;
  }
};

Matrix

Parse 2D matrices (e.g. PAM)

read(url)

Parses an url an calls your parse method with the returned body.

MParser.read("https://cdn.rawgit.com/greenify/biojs-io-matrix/master/test/data/pam_250", function(err, model) {
	// model is the parsed url
});

If callback is undefined, read returns a promise.

var p = MParser.read("https://cdn.rawgit.com/greenify/biojs-io-matrix/master/test/data/pam_250");
// ...
p.then(function(model) {
	// model is the parsed url
}, function(err){
	console.error("err happened during downloading", err);
});

Read a entire file

var matrix = MParser.parse("A 1\nB 1 2");
> { A: { A: 1, B: 1 },
  B: { A: 1, B: 2 } }

Line by line

var mParser = new MParser();
mParser.parseLine("A 1");
mParser.parseLine("B 2 3");
var matrix = mParser.buildMatrix();

Save a matrix

var matrix = MParser("A 1\nB 1 2");
var out = mParser.export(matrix);
> 'A\t1\nB\t1\t2'

Or you can use objects

var matrix = new MParser("A 1\nB 1 2");
matrix.buildMatrix(); \\ returns the 2D array
var out = matrix.export();
> 'A\t1\nB\t1\t2'

BLAST

BLAST parser

It parses the XML output of BLAST. You can activate the XML output by adding the -outfmt 5 flag to your BLAST program.

More info about BLAST

Getting Started

Install the module with: npm install biojs-io-blast

var blast = require('biojs-io-blast');
blast.read("http://files.biojs.net/blast/examples/syne1.xml", function(data){
	console.log("blast object", data);
});

Play on JSBin

Documentation

BLAST: short recap (with the output format)

.read(url)

Parameter: URL of a BLAST output (in xml) Type: String Example: http://files.biojs.net/blast/examples/syne1.xml

Parameter: Callback or promise Type: function

Downloads the XML BLAST file and parses it to JSON.

blast.read('http://files.biojs.net/blast/examples/syne1.xml', function(err, data){
	console.log("blast object", data);
});

If you don't specify a callback, it will return a promise.

var p = blast.read("https://raw.githubusercontent.com/greenify/biojs-io-blast/master/test/dummy/single.xml");
p.then(function(model) {
    // model is the parsed url
}, function(err){
    console.error("err happened during downloading", err);
});

As downloading the file is asynchronous, it will call your callback with the resulting data object.

.parse(str)

Parameter: XML Output of BLAST Type: String Example: <BlastOutput><BlastOutput_program>blastp</BlastOutput_program></BlastOutput>

Parses the BLAST XML output to JSON.

blast.parse('<BlastOutput><BlastOutput_program>blastp</BlastOutput_program></BlastOutput>');

CLI version

Install it globally npm install -g biojs-io-blast and then you can pipe directly into the parser.

cat <blast-file.xml> | biojs-blast

Output

• there can be multiple iterations
• there can be multiple per hits per iteration
• there can be multiple HSPS per hit

Full example

Official BLAST XML spec

{
    "program": "blastp",
    "version": "BLASTP 2.2.29+",
    "reference": "Stephen F. Altschul, Thomas L. Madden, Alejandro A. Schäffer, Jinghui Zhang, Zheng Zhang, Webb Miller, and David J. Lipman (1997), \"Gapped BLAST and PSI-BLAST: a new generation of protein database search programs\", Nucleic Acids Res. 25:3389-3402.",
    "db": "/home/xsebi/tmp/blast/swiss",
    "query-ID": "Query_1",
    "query-def": "sp|Q8NF91|SYNE1_HUMAN Nesprin-1 OS=Homo sapiens GN=SYNE1 PE=1 SV=4",
    "query-len": "8797",
    "param": {
        "matrix": "BLOSUM62",
        "expect": "10",
        "gap-open": "11",
        "gap-extend": "1",
        "filter": "F"
    },
    "iterations": [
        {
            "iter-num": "1",
            "query-ID": "Query_1",
            "query-def": "sp|Q8NF91|SYNE1_HUMAN Nesprin-1 OS=Homo sapiens GN=SYNE1 PE=1 SV=4",
            "query-len": "8797",
            "hits": [
                {
                    "num": "1",
                    "id": "gnl|BL_ORD_ID|140988",
                    "def": "gi|425906075|sp|Q8NF91.4|SYNE1_HUMAN RecName: Full=Nesprin-1; AltName: Full=Enaptin; AltName: Full=Myocyte nuclear envelope protein 1; Short=Myne-1; AltName: Full=Nuclear envelope spectrin repeat protein 1; AltName: Full=Synaptic nuclear envelope protein 1; Short=Syne-1 [Homo sapiens]",
                    "accession": "140988",
                    "len": "8797",
                    "hsps": [
                        {
                            "num": "1",
                            "bit-score": "17954.1",
                            "score": "46598",
                            "evalue": "0",
                            "query-from": "1",
                            "query-to": "8797",
                            "hit-from": "1",
                            "hit-to": "8797",
                            "query-frame": "0",
                            "hit-frame": "0",
                            "identity": "8797",
                            "positive": "8797",
                            "gaps": "0",
                            "align-len": "8797",
                            "qseq": "MATSRGASRCPR...",
                            "hseq": "MATSRGASRCPR...",
                            "midline": "MATSRGASRCPR..."
                        }
                    ]
            "stat": {
                "db-num": "459767",
                "db-len": "171814008",
                "hsp-len": "143",
                "eff-space": "917906647858",
                "kappa": "0.041",
                "lambda": "0.267",
                "entropy": "0.14"
            }
        }
    ]
}

Newick Parser

Just call method parse_newickk(string) for parsing a newick string into JSON.

Biojs.io.newick.parse_newick('((A,B),C)');

Call the method parse_nhx(string) for parsing an extended newick formats into JSON.

Biojs.io.newick.parse_nhx('((A,B),C)');

Example tree:

Newick format:

(A:0.1,B:0.2,(C:0.3,D:0.4)E:0.5)F

Converted to JSON:

{name : "F",
  children: [
    {name: "A", branch_length: 0.1},
    {name: "B", branch_length: 0.2},
    {
      name: "E",
      length: 0.5,
      children: [
        {name: "C", branch_length: 0.3},
        {name: "D", branch_length: 0.4}
      ]
    }
  ]
}