concept · intermediate · partly written
What the data object is, layer by layer
The file formats and tables that each omics layer lands in, who defines each, and which are standards and which are one group's choice that became practice.
Standard, practice or one group's choice
Most data objects in daily use began as one project's design and became practice: FASTQ, SAM and BAM, VCF and AnnData. A smaller set is written by a standards body or consortium: mzML and mzTab-M for mass spectrometry, SBML for models, OME-NGFF for images. No source opened measures how often any of them is followed. 12345678
Reported, not yet replicated
Genomics: reads, alignments, variants, pangenome more
FASTQ was in use for years without a formal definition, in three incompatible variants, until a 2009 community write-up (abstract only read). SAM and BAM came from the SAMtools group and are keyed on coordinates of one reference genome build. VCF was built for the 1000 Genomes Project and then adopted elsewhere (abstract only read). A pangenome is a sequence graph in which nodes are DNA strings and each assembly is a haplotype path; the first release of the Human Pangenome Reference Consortium holds 47 phased diploid assemblies, and its minigraph build records only variants of 50 bases or more. CRAM and the GFA specification were not opened. 1239
Reported, not yet replicated
Transcriptomics: matrices, archives, single-cell objects more
A bulk or array experiment lands as a count or intensity matrix. GEO organises submissions as platform, sample and series; the MIAME rules set what content must be reported, not the technical file format. For single cells, AnnData holds a matrix X plus annotation tables for cells and features, embeddings and layers, stored as HDF5 or Zarr. It comes from the Scanpy group, not from a standards body. MuData keeps one AnnData per modality and allows a modality to be missing for some cells. The CELLxGENE schema 6.0.0, enforced on that portal's own submissions, requires raw integer counts for listed assays, a Cell Ontology term or the word unknown for cell type, and one species and one Ensembl release per dataset. 10114121314
Reported, not yet replicated
Proteomics: spectra, results, proteoforms, samples more
mzML (HUPO-PSI) records spectra with a controlled vocabulary checked by validators rather than a rigid schema. The mzTab family reports identification and quantification results; protein identity in them is inferred, not observed. ProForma 2.0 writes a modified protein form with modifications in brackets after the residue, using terms from Unimod, PSI-MOD, RESID, XL-MOD and GNO, and allows unlocalised modifications and mass shifts. SDRF-Proteomics, adapted from the transcriptomics MAGE-TAB, links each sample to its data file; the official PSI specification is dated 24 May 2023, and no compliance rate for it was found. 5151617
Reported, not yet replicated
Epigenomics and metabolomics, briefly more
Chromatin accessibility is stored as peak matrices or as gene-activity scores in the same containers as RNA; no format of its own turned up. In the scIB integration benchmark the gene-activity scores were the weakest feature space. For metabolomics, mzTab-M has a mandatory controlled vocabulary and, by design, allows ambiguity in the identification of molecules, so one entry can stand for several candidate metabolites. 186
Reported, not yet replicated
Spatial: SpatialData more
SpatialData holds five element types: Images, Labels (raster masks), Points, Shapes and Tables. Rasters follow the OME-NGFF standard in Zarr; points and shapes are stored as Parquet. Rasters are a defined community standard; the rest is one group's design. 19208
Reported, not yet replicated
What is lost when identifiers are mapped more
Measured loss is rare. For protein inference, one human-cell dataset of five runs with 293 to 338 proteins each had about 55% with unique evidence, 21 to 28% in groups that cannot be told apart, and about 20% non-conclusive; the dataset is small and shallow, so these are not proteome rates. This atlas found no measured percentage lost between Ensembl, RefSeq, HGNC and UniProt. 21
Reported, not yet replicated
Plain answer
Why it matters
Real numbers
First human pangenome release 2026-11-18
47 phased diploid assemblies as a sequence graph. 9
Reported, not yet replicated
Protein inference ambiguity, one dataset 2026-11-18
About 55% of proteins with unique evidence, 21 to 28% in indistinguishable groups, about 20% non-conclusive (five runs of 293 to 338 proteins). 21
Reported, not yet replicated
ProteomeXchange 2026-11-18
64,330 datasets submitted through June 2025, 44,248 of them public; PRIDE holds 77% (the consortium's own report). 15
Reported, not yet replicated
Still open here
- SummarizedExperiment, GFA, CRAM, SBGN, imzML, OME-TIFF and the CODEX and MIBI formats were not opened.
- What percentage of genes or proteins is lost between Ensembl, RefSeq, HGNC and UniProt? See questions/what-is-the-identifier-mapping-loss-between-gene-and-protein-databases.
Not yet written: Equation.
Linked from
- How omics layers are joined: by identifier, sample, cell, position, latent space and graph
- Imageromics (Amit's word) and the technologies it covers
- How a multimodal single-cell or spatial data object ties its parts together
- Proteomics standards and archives: mzML, mzTab, ProForma, SDRF and ProteomeXchange
- What a missing value means in each omics layer
- Global Alliance for Genomics and Health
- HUPO Proteomics Standards Initiative
- scverse
- What fraction of genes and proteins is lost or split when mapping between Ensembl, RefSeq, HGNC and UniProt?
Sources
- The Sanger FASTQ file format for sequences with quality scores, and the Solexa/Illumina FASTQ variants. Nucleic Acids Research (2010). paper; opened 2026-11-18 all records citing this
- Li H, Handsaker B, Wysoker A, Fennell T, et al.. The Sequence Alignment/Map format and SAMtools. Bioinformatics (2009). paper; opened 2026-11-18 all records citing this
- The variant call format and VCFtools. Bioinformatics (2011). paper; opened 2026-11-18 all records citing this
- Wolf FA, Angerer P, Theis FJ. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol (2018). paper; opened 2026-11-18 all records citing this
- Martens L, Chambers M, Sturm M, Kessner D, et al.. mzML--a community standard for mass spectrometry data. Mol Cell Proteomics (2011). paper; opened 2026-11-18 all records citing this
- Hoffmann N, Rein J, Sachsenberg T, Hartler J, et al.. mzTab-M: A Data Standard for Sharing Quantitative Results in Mass Spectrometry Metabolomics. Anal Chem (2019). paper; opened 2026-11-18 all records citing this
- Hucka M, Bergmann FT, Chaouiya C, Dräger A, et al.. The Systems Biology Markup Language (SBML): Language Specification for Level 3 Version 2 Core Release 2. J Integr Bioinform (2019). paper; opened 2026-11-18 all records citing this
- Moore J, Allan C, Besson S, Burel JM, et al.. OME-NGFF: a next-generation file format for expanding bioimaging data-access strategies. Nat Methods (2021). paper; opened 2026-11-18 all records citing this
- Liao WW, Asri M, Ebler J, Doerr D, et al.. A draft human pangenome reference. Nature (2023). paper; opened 2026-11-18 all records citing this
- Brazma A, Hingamp P, Quackenbush J, et al.. Minimum information about a microarray experiment (MIAME)—toward standards for microarray data. Nature Genetics (2001). paper; opened 2026-11-18 all records citing this
- Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Research (2002). paper; opened 2026-11-18 all records citing this
- On-disk format (anndata documentation).. web, not peer reviewed; opened 2026-11-18 all records citing this
- Bredikhin D, Kats I, Stegle O. MUON: multimodal omics analysis framework. Genome Biol (2022). paper; opened 2026-11-18 all records citing this
- CELLxGENE single-cell curation schema 6.0.0.. web, not peer reviewed; opened 2026-11-18 all records citing this
- Deutsch EW, Bandeira N, Perez-Riverol Y, Sharma V, et al.. The ProteomeXchange consortium in 2026: making proteomics data FAIR. Nucleic Acids Res (2026). paper; opened 2026-11-18 all records citing this
- ProForma 2.0 (HUPO-PSI repository).. web, not peer reviewed; opened 2026-11-18 all records citing this
- SDRF-Proteomics, sample and data relationship format (HUPO-PSI).. organisation-page, not peer reviewed; opened 2026-11-18 all records citing this
- Luecken MD, et al.. Benchmarking atlas-level data integration in single-cell genomics. Nature Methods (2022). paper; opened 2026-11-18 all records citing this
- Design document for SpatialData.. web, not peer reviewed; opened 2026-11-18 all records citing this
- Marconato L, Palla G, Yamauchi KA, Virshup I, et al.. SpatialData: an open and universal data framework for spatial omics. Nat Methods (2025). paper; opened 2026-11-18 all records citing this
- Prieto G, Aloria K, Osinalde N, Fullaondo A, et al.. PAnalyzer: a software tool for protein inference in shotgun proteomics. BMC Bioinformatics (2012). paper; opened 2026-11-18 all records citing this
Where it sits on the map
History of this record (6)
This record has been changed 6 times, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.
- Status
- partly written
- Last updated
- 2026-11-19
- Record
- content/concepts/omics-data-objects-by-layer.json