tximeta provides a set of functions for conveniently working with
metadata for transcript quantification data in Bioconductor. The
tximeta() function imports quantification data from salmon or
related quantifiers, and returns a
SummarizedExperiment
object. tximeta works natively with
salmon,
alevin,
piscem-infer, and
oarfish,
but can easily be configured to work with any transcript
quantification tool.
If tximeta() recognizes the reference transcripts used
for quantification, it will automatically download relevant
information about the location of the transcripts in the correct genome.
These actions happen in the background without requiring any extra
effort or information from the user.
This metadata is attached to the SummarizedExperiment in the
metadata() and rowRanges() or rowData() slots.
For a list of the reference transcriptomes supported by tximeta(),
see the "Pre-computed digests" section of the vignette in the
Get started tab.
Note that in tximeta documentation, we call the computed identifier
for the reference transcriptome a "digest" or sometimes a "checksum",
which is produced by hash function(s) employed by upstream software.
Further steps are also facilitated, e.g. summarizeToGene(), addIds(),
or even retrieveCDNA() (the transcripts used for quantification) or
retrieveDb() (the correct TxDb or EnsDb to match the
quantification data).
The key idea behind tximeta is that Salmon, alevin, and
piscem-infer propagate a hash value
summarizing the reference transcripts into each quantification
directory it outputs. tximeta can be used with other tools as long
as the
hash of the transcripts
is also included in the output directories. See customMetaInfo
argument of tximeta() for more details.
In the Bioconductor 3.22 release (October 2025), tximeta's long read import
pipeline was updated to support mixed reference transcripts, where
oarfish is used to quantify against
a combination of --annotated (e.g. GENCODE, Ensembl) and --novel
(e.g. de novo assembled) reference transcripts. New functions importData(),
inspectDigests(), and updateMetadata() facilitate import and automatic
metadata attachment across both reference sets. Support for salmon mixed
reference workflows is also available via the mixedDigest argument of
importData(), using a companion Snakemake rule to record per-sub-index digests
at index time.
See the mixed reference transcripts
section of the vignette for details.
A reference for tximeta package is:
Michael I. Love, Charlotte Soneson, Peter F. Hickey, Lisa K. Johnson, N. Tessa Pierce, Lori Shepherd, Martin Morgan, Rob Patro. "Tximeta: reference sequence checksums for provenance identification in RNA-seq" PLOS Computational Biology (2020) doi: 10.1371/journal.pcbi.1007664
We would love to hear your feedback. Please post to Bioconductor support site for software usage help or post an Issue on GitHub, for software development questions.
tximeta was developed as part of NIH NHGRI R01-HG009937.
tximeta was also supported by the Chan Zuckerberg Initiative as part of the EOSS grants.