Sample and Data Relationship Format for Proteomics
The SDRF-Proteomics file format describes the sample characteristics and the relationships between samples and data files. The file format is a tab-delimited one where each ROW corresponds to a relationship between a Sample and a Data file (and channel in the context of labelled samples), each column corresponds to an attribute/property of the Sample and the value in each cell is the specific value of the property for a given Sample.
The SDRF-Proteomics is divided into three main blocks:
- characteristics[...]: These are properties of the sample of origin.
- comment[...]: These are the data properties.
- factor value[...]: These are the variables under study.
The SDRF columns MUST starts with the source name which is the sample accession. After the sample accession all the columns correspond to the sample characteristics, for example (characteristics[organism]), until the assay name column which starts the Data file section.
The Data properties section (comment) starts with the assay name which is the Data file accession. After the assay name the following properties (comment) are mandatory for SDRF-Proteomics:
-
comment[label]: The label is the channel used in multiplexed experiments (e.g, TMT126 - check the documentation for the labelled methods). If the sample is not label free or the experiments haven't used any multiplex analytical method, the value MUST BE
label free sample. - comment[fraction identifier]: The fraction identifier is a unique identifier for each Data file. Fraction identifiers help to identify any type of Fractionation method including: High-performance liquid chromatography, Isoelectric focusing or Off-gel electrophoresis.
The values properties within the main three blocks chracteristics, comment and factor value, MUST be ontology terms that can be found in the OLS ontology service. The SDRF-Proteomics file format allows researchers and submitters to go from a simple file format like:
| source name | characteristics[organism] | characteristics[organism part] | characteristics[biological replicate] | assay name | comment[technical replicate] | comment[fraction identifier] | comment[label] | factor value[organism part] |
|---|---|---|---|---|---|---|---|---|
| Sample-1 | homo sapiens | heart | 1 | ms_run 1 | 1 | 1 | label free sample | heart |
| Sample-2 | homo sapiens | liver | 1 | ms_run 2 | 1 | 1 | label free sample | liver |
The previous example, only contains the minimum information a researcher needs to understand the Sample and Data file relationship. The factor value is used to define which characteristic from the sample is under study (e.g. organism part). The example can be read as: Two different label-free samples (one from the liver and one from the heart in human) with no fractionation are compared.
With the following properties source name, characteristics[biological replicate], assay name, comment[technical replicate], comment[label], comment[fraction identifier], factor value[sample property], the submitter can annotate the relation between the sample and the data file (and label channel in the context of multiplexed experiment). However, more metadata is needed in order to understand the Sample source, the data acquisition protocol etc.
ProteomeXchange partners have defined the SDRF-Proteomics templates; a group of guidelines and checklists of minimum sample metadata requested for different types of experiment. For example, for Human datasets the following metadata MUST be provided:
- source name: Sample identifier
- characteristics[organism]: Organism
- characteristics[ancestry category]: Ancestry category
- characteristics[age]: Age of the individual, the age should be formatted as the specification recommends
- characteristics[sex]: Sex
- characteristics[disease]: Disease under study
- characteristics[organism part]: Organism part
- characteristics[cell type]: Cell type
- characteristics[individual]: Unique identifier for the individual or patient
- characteristics[biological replicate]: Biological replicate accession. These variable is related with the factor value.
As mentioned before, for each Data file the following metadata is required:
- assay name: Identifier of the file
- comment[label]: The labelled used in the experiment. Some tools recognise it as channel.
- comment[fraction identifier]: Fraction identifier accession. If no fractionation is used, all the values should be
1. - comment[technical replicate]: Technical replicate accession.
- comment[data file]: Data file name, including the extension.
Additionally, the ProteomeXchange request for every dataset to provides the following properties:
- comment[cleavage agent details]: Enzymes and cleavage agents use in the experiment
- comment[instrument]: MS instrument used to acquire the data.
Templates are designed to capture the minimum metadata at sample and data level, depending on the type of the experiment. For example for cell line datasets, the characteristics[cell line] MUST be provided.
The ProteomeXchange consortium RECOMMEND providing additional information to each RAW file. These additional information include:
- comment[modification parameters]: Post-translational modifications searched in the experiment.
- comment[fragment mass tolerance]: Fragment tolerances
- comment[precursor mass tolerance]: Precursor tolerance
