fgbio

A set of tools to analyze genomic data with a focus on Next Generation Sequencing.

This readme document is mostly for developers/contributors and those attempting to build the project from source. Detailed user documentation is available on the project website including tool usage and documentation of metrics produced. Detailed developer documentation can be found here.

• Quick Installation
• Goals
• Overview
• List of tools
• Building
• Command line
• Include fgbio in your project
• Contributing
• Authors
• License
• Sponsorship

Quick Installation

The conda package manager (configured with bioconda channels) can be used to quickly install fgbio:

conda install fgbio

To install fgbio without extra dependencies (e.g. R), use the command:

conda install fgbio-minimal

Goals

There are many toolkits available for analyzing genomic data; fgbio does not aim to be all things to all people but is specifically focused on providing:

• Robust, well-tested tools.
• An easy to use command-line.
• Clear and thorough documentation for each tool.
• Open source development for the benefit of the community and our clients.

Overview

Fgbio is a set of command line tools to perform bioinformatic/genomic data analysis. The collection of tools within fgbio are used by our customers and others both for ad-hoc data analysis and within production pipelines. These tools typically operate on read-level data (ex. FASTQ, SAM, or BAM) or variant-level data (ex. VCF or BCF). They range from simple tools to filter reads in a BAM file, to tools to compute consensus reads from reads with the same molecular index/tag. See the list of tools for more detail on the tools

List of tools

For a full list of available tools please see the tools section of the project website.

Below we highlight a few tools that you may find useful.

• Tools for working with Unique Molecular Indexes (UMIs, aka Molecular IDs or Molecular Barcodes):
  • Annotate/Extract Umis from read-level data: FastqToBam, AnnotateBamWithUmis, ExtractUmisFromBam, and CopyUmiFromReadName.
  • Manipulate read-level data containing Umis: CorrectUmis, GroupReadsByUmi, CallMolecularConsensusReads, CallDuplexConsensusReads, and FilterConsensusReads.
  • Collect metrics and review consensus reads: CollectDuplexSeqMetrics and ReviewConsensusVariants.
• Tools to manipulate read-level data:
  • Fastq Manipulation: FastqToBam, ZipperBams, and DemuxFastqs (see [fqtk][fqtk-link], our rust re-implementation for sample demultiplexing).
  • Filter, clip, randomize, sort, and update metadata for read-level data: FilterBam, ClipBam, RandomizeBam, SortBam, SetMateInformation and UpdateReadGroups.
• Tools for quality control assessment:
  • Detailed substitution error rate evaluation: ErrorRateByReadPosition.
  • Sample pooling QC: [EstimatePoolingFractions]: fgbio-estimatepoolingfractions-link.
  • Splice-aware insert size QC for RNA-seq libraries: EstimateRnaSeqInsertSize.
• Tools for adding or manipulating alternate contig names:
  • Extract contig names from an NCBI Assembly Report: CollectAlternateContigNames.
  • Update contig names in common file formats: UpdateFastaContigNames, UpdateVcfContigNames, UpdateGffContigNames, UpdateIntervalListContigNames, UpdateDelimitedFileContigNames.
• Miscellaneous tools:
  • Pick molecular indices (ex. sample barcodes, or molecular indexes): PickIlluminaIndices and PickLongIndices.
  • Find technical/synthetic, or switch-back sequences in read-level data: FindTechnicalReads and FindSwitchbackReads.
  • Make synthetic mixture VCFs: MakeMixtureVcf and MakeTwoSampleMixtureVcf.

Building

Cloning the Repository

Git LFS is used to store large files used in testing fgbio. In order to compile and run tests it is necessary to install git lfs. To retrieve the large files either:

1. Clone the repository after installing git lfs, or
2. In a previously cloned repository run git lfs pull once

After initial setup regular git commands (e.g. pull, fetch, push) will also operate on large files and no special handling is needed.

To clone the repository: git clone https://github.com/fulcrumgenomics/fgbio.git

Running the build

fgbio is built using sbt.

Use sbt assembly to build an executable jar in target/scala-2.13/.

Tests may be run with sbt test.

Java SE 8 is required.

Command line

java -jar target/scala-2.13/fgbio-<version>.jar to see the commands supported. Use java -jar target/scala-2.13/fgbio-<version>.jar <command> to see the help message for a particular command.

Include fgbio in your project

You can include fgbio in your project using:

"com.fulcrumgenomics" %% "fgbio" % "1.0.0"

for the latest released version or (buyer beware):

"com.fulcrumgenomics" %% "fgbio" % "0.9.0-<commit-hash>-SNAPSHOT"

for the latest development snapshot.

Contributing

Contributions are welcome and encouraged. We will do our best to provide an initial response to any pull request or issue within one-week. For urgent matters, please contact us directly.

Authors

• Tim Fennell (maintainer)
• Nils Homer (maintainer)

License

fgbio is open source software released under the MIT License.

Sponsorship

Become a sponsor

As a free and open source project, fgbio relies on the support of the community of users for its development. If you work for an organization that uses and benefits from fgbio, please consider supporting fgbio. There are different ways, such as employing people to work on fgbio, funding the project, or becoming a sponsor to support the broader ecosystem. Please contact@fulcrumgenomics.com to discuss.

Sponsors

Sponsors provide support for fgbio through direct funding or employing contributors. Public sponsors include:

                 

The full list of sponsors supporting fgbio is available in the sponsor page.