Birdie

birdie is a bioinformatics module for Deno that aims to evolve existing JavaScript code to WebAssembly-powered variants for increased performance. Currently it’s in very early stages of development and all features are considered unstable and experimental.

Features

String Edit Distance

Import and initialize the WebAssembly module:

import * as distance from "https://deno.land/x/birdie/distance/mod.ts";

await distance.initWasm();

Code using this module will need permissions to fetch .wasm binaries, which requires --allow-net=deno.land.

Hamming Distance

const x = "GTCTGCATGCG";
const y = "TTTAGCTAGCG";

const hammingDistance = distance.hamming(x, y);
console.log(hammingDistance); // 5n (BigInt)

Levenshtein distance (Wagner-Fischer Algorithm)

const levenshteinDistance = distance.levenshtein("kitten", "sitting");
console.log(levenshteinDistance); // 3

Levenshtein distance (Myers’ Algorithm)

const myersDistance = distance.myers("kitten", "sitting");
console.log(myersDistance); // 3

Sequence Analysis

import * as sequence from "https://deno.land/x/birdie/sequence/mod.ts";

await sequence.initWasm();

const gcRatio = sequence.gcContent("GATATACA");

console.log(gcRatio); // 0.25

const gc3Ratio = sequence.gc3Content("GATATACA");

console.log(gc3Ratio); // 0.67

Pattern Matching

shift_and

import * as patternMatching from "https://deno.land/x/birdie/pattern_matching/mod.ts";

await patternMatching.initWasm();

const pattern = "AAAA";
const text = "ACGGCTAGAAAAGGCTAG";

const patternStartingPosition = patternMatching.shiftAnd(pattern, text);
console.log(patternStartingPosition); // 8

Contributing

Any contributions are welcome and much appreciated!