Experiments

When submitting data to the IDR, you should provide 3 basic files:

1. study file describing the overall study and the experiments that were performed. An experiment is a considered to be a discrete piece of work, usually with its own set of protocols
2. assay file(s) describing the images in each experiment
3. processed data file(s) containing summary results and/or a ‘hit' list for each experiment

All files should be in tab-delimited text format. Templates are provided but can be modified to suit your experiment. Add or remove columns from the templates as necessary. The templates can be opened in Excel, Open Office etc.

This file should contain a general description of the study, and then list the experiments(s) performed, along with the protocols describing how each experiment was set up, imaged and the data analyzed.

1a. Study description:

• Study title (mandatory)
• Study description (mandatory)
• Experiment publication (PubMed ID, if applicable)
• Primary contact information (mandatory)

1b. Study experiments:

For each experiment the following information should be included:

• Experiment description – a brief description of the aims or contents of the experiment (mandatory)
• Protocols –
  • cell growth
  • treatment (if applicable)
  • image acquisition and feature extraction
  • analysis – the analysis pipeline to go from the image data to the results
• Phenotypes identified – list any cellular phenotypes identified e.g. large nucleus, elongated cell, delay of mitosis, localization of protein
• Name of the assay file (mandatory)
• Name of the processed data file (mandatory)
• Description of each column in the processed data file (mandatory)

If there is more than one experiment in the study, then copy and past the previous ‘experiment block' of text so it can be filled in again for the next experiment.

Each experiment should have an assay file describing the samples and the images of them. In this file, each image should have its own row, and it is almost like a time line, with the starting sample described first (left hand side of the row) and then moving through treatments to the sample (referencing the protocol names described in study file), then the imaging assay, then the data analysis and finally listing the file with the results in it. In more detail, working from left to right the row for each image should contain:

• sample information (e.g. species, cell line, genotype, organism part),
• growth, treatment (if applicable), and image acquisition protocols should be listed,
• assay name. Assay names are for each imaging ‘assay' but there may be several images from an assay, e.g. several images from the same slide or organism part, in which case they should all be given the same assay name to group them together.
• Experimental conditions. Experimental conditions are any important variables in the experiment e.g. were the cells treated with different compounds, the protein targeted for labeling varies, or the organism part is different in each image.
• Dataset name. This is usually the same as the assay name, but if you want to give it a different ‘display name' in the image repository you can specify it here.
• Image File Path – the path to the file on the file system (starting from the folder which contains all the image files)
• Image File name – the exact name of the image file
• Channels – the label and what it targets, separated by a colon. Each pair of labels and visualized object is separated by a semi-colon. E.g. DAPI:nuclei;TRITC:HA_Flag tagged protein.
• Data processing protocols – list the name of the data analysis protocol described in the study file.
• Processed data – the name of the file containing the processed data/summary results.
• Comment[] columns can be added at any point and values in the square brackets created as needed (although IDR curators may edit the values so that they are similar across different studies) e.g. Comment [Image File Type] to say if the file is a raw or derived image file.

Each experiment should have a processed data file that contains summary information about the results found. This may be a table from the associated publication. This file can contain information relating to each image or be values calculated from several images.

The information in the processed data should be linkable to the assay file in some way e.g. dataset name

The contents of each column in the processed file should be described in the study file so that it is clear what the values are.

Lists of terms that can be used in the study and library file can be found in this google doc.

Example files can be found in https://github.com/IDR/idr-metadata. If you are not familiar with github, it is a way of storing versions of files. You can browse the files through the github interface. If you want to download a file you can click on the ‘Raw' button on the top right hand side above the file preview and then do File -> Save Page As in your browser, or if you are familiar with github you can clone the repository. There are files in addition to the study, library and processed files in the repository but you don't need to provide these.

Examples of non-screen studies include:

• idr0018 - histopathology of gene knockouts
• idr0021 - protein localization using 3D-SIM
• idr0023 - protein localization using dSTORM
• idr0027 - 3D-tracking of tagged chromatin loci
• idr0032 - in situ hybridzation of RNA

Email idr@openmicroscopy.org with any questions.