DnaDesign

DnaDesign is a Go/Lua project creating tools for automated genetic design, spanning from the lowest levels of DNA sequence manipulation to higher level functionality.

• Practical: DnaDesign tooling is meant to be used by practitioners of synthetic biology.

• Modern: DnaDesign is at the bleeding edge of technology. We are happy to adopt the newest advancements in synthetic biology, LLMs, and more to get our tools in the hands of people who need it.

• Ambitious: DnaDesign's goal is to be the most complete, open, and well used collection of computational synthetic biology tools ever assembled. If you like our dream and want to support us please star this repo, request a feature, or open a pull request.

Documentation

• Go Library Documentation

Go/Lua

DnaDesign is implemented with both lua and go versions. They should work identically, and anything that does not work identically is considered a bug. The original codebase was written in Go and we currently have production software running the Go version, but most development now is done in the lua version. This is because, most of the time, just a quick script needs to be written, and lua is fantastic for this. Unlike python, lua can be trivally embedded as a scripting language, allowing the DnaDesign library to be embedded anywhere and scripted by LLMs.

DnaDesign is written in teal, a statically typed lua language, and compiled to lua. A lua amalgamation is provided to embed into any system.

Repo organization

On the highest level:

• lua contains core functionality as a lua library.
• go/lib contains core functionality as a go library.
• go/external contains functions to work with external bioinformatics command-line interfaces.

Detailed repo organization

• go/lib contains the core DnaDesign library, with nearly all functionality, all in idiomatic Go with nearly no dependencies.
  • go/lib/bio contains biological parsers for file formats including genbank, fasta, uniprot, fastq, slow5, sam, and pileup files.
  • go/lib/align contains Needleman-Wunsch and Smith-Waterman alignment functions, as well as the mash similarity algorithm.
  • go/lib/clone contains functions for simulating DNA cloning, including restriction digestion, ligation, and GoldenGate assembly.
  • go/lib/fold contains DNA and RNA folding simulation software, including the Zuker and LinearFold folding algorithms.
  • go/lib/primers contains DNA primer design functions.
    • go/lib/primers/pcr contains PCR simulation functions.
  • go/lib/seqhash contains the Seqhash algorithm to create universal identifiers for DNA/RNA/protein.
  • go/lib/synthesis contains various functions for designing synthetic DNA.
    • go/lib/synthesis/codon contains functions for working with codon tables, translating genes, and optimizing codons for expression.
    • go/lib/synthesis/fragment contains functions for optimal GoldenGate fragmentation.
    • go/lib/synthesis/fix contains functions for fixing proteins in preparation for synthesis.
  • go/lib/transform contains basic utility functions for transforming DNA, like reverse complementation.
• external contains integrations with external bioinformatics software, usually operating on the command line.
  • go/external/minimap2 contains a function for working with minimap2 with Go.
  • go/external/samtools contains a function for generating pileup files using samtools with Go.
  • go/external/bcftools contains GenerateVCF to generate a VCF file from sam alignments using bcftools with Go.

Contributing

Write good, useful code. Open a pull request, and we'll see if it fits!

License

• MIT

Sources

The lua version of dnadesign is completely self-contained.

There are a few pieces of "complete" Go software that we have directly integrated into the source tree (with their associated licenses). These projects don't receive updates anymore (or just bug fixes with years between). In particular, lib has most of these, since we intend to have as few dependencies as possible in lib. The integrated projects include the following.

• svb in lib/bio/slow5/svb
• intel-cpuid in lib/bio/slow5/svb/intel-cpuid
• wordwrap in lib/bio/genbank
• errgroup in lib/bio/

Other

DnaDesign is a fork of Poly at commit f76bf05 with a different mission focus.

Changelog

All notable changes to this project will be documented here.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

• Removed python version of dnadesign (too hard to maintain)
• Adds uniref parser #107
• Fixes iso-8859-1 error in reading uniref data dumps #106
• Updates uniprot parser to read IDs #104
• Fixes RecursiveFragment to not add flanks to the initial input #102
• Fixes add flank bug, releases new version of python lib #101
• Adds feature for adding flanks to RecursiveFragment. #100
• Adds cloning and recursion functions to python. #96
• Adds recursive fragmentation. #92
• Updated megamash documentation to be more specific. #91
• Adds automatic python documentation generation. #88
• Adds genbank parsing to python package. Release version 0.1.5 of dnadesign python. #87
• Adds fastq parsing to python package. Releases version 0.1.4 of dnadesign python. #86
• Integrated errgroup into source tree #84
• Added kmer detection for ligation events in cloning and removed enzyme manager #83
• Added option for linear ligations #82
• Added minimal python packaging #81
• Greatly simplified the Ligate function #77
• Updated barcoding functions to handle edge case of hanging-edge barcodes #74
• Updated megamash to use int instead of uint for minimal Kmer counts (so you can use -1) #73
• Added bcftools to external #72
• Fixed bug in PCR where certain matching primers wouldn't create any amplicons #71
• Updated seqhash2 to use base58 rather than base64 #69
• Updated dual barcodes primer sets to be created without csv files #67
• Added workers to bio as a way to process data #62
• Improved megamash efficiency and added []Match JSON conversion #61
• Added barcoding functionality for sequencing reads #59
• Added the megamash algorithm #50
• Changed parsers to return values instead of pointers. Added some sequencing utils #49
• Added minimap2 and samtools(pileup) integrations in external #46
• Added sam parser #5
• Added the LinearFold folding algorithms #38
• Added Get function to uniprot for getting a single uniprot xml from online #37
• Removed murmur3 in favor of crc32 for mash #33
• Patch start codon problems #32
• Added tests for OpenBSD #31
• Removed a large number of unneeded dependencies #28
• Added full rebase text database into tree 27b41fb
• Updated uniprot to be a standardized parser #22
• Purged unsupported gff parser #23
• Moved library to lib directory #21
• Fixed issue with JSON codon tables #4
• Added Seqhash v2 #3
• Added lowercase methylation options during cloning #2
• Standardized parsers with generics #1