How to upload TMA data (anonymous users)

The process of uploading a dataset consists of two steps which are described in the sections below. Briefly, the first step is to supply a file containing marker values (typically TMA protein expression scores); the second step involves selecting options to be applied during the upload and providing information about the dataset (e.g. a title). When the dataset is uploaded, it is added to a queue for processing. You can then either wait until processing completes (usually a few moments depending on file size), or bookmark the page in your web browser for a later visit. Once processing has completed successfully, the dataset is ready for analysis. The help section data requirements - marker scores gives information on how to organise your TMA data for successful uploading.

You can upload data using the Add dataset button on the top right of every page (figure 1).

Figure 1: The Add dataset button (highlighted in red) is present on the top right of every page.

Warning: Datasets uploaded by anonymous users and associated results will be deleted after six months. Please be sure to save any data you need within the six month period.

Stage 1: File validation

From the home page, click on Start workflow. You will now be on a page titled Upload a dataset.

To begin, click on the button in the middle of the page labelled Upload data (figure 2). The add dataset form will then appear (figure 3).

You can also access the add dataset form from any page on the site by using the Add dataset button in the top right (figure 1, above).

To upload a dataset, click the upload data button

Figure 2: Click the Upload data button (highlighted in red) to begin uploading TMA data

Figure 3: The add dataset form, used to upload and annotate data.

With the add dataset form open (figure 3, above) click the button labelled Upload a file. You will then be asked to choose a file from your computer. Once you have selected a file, click OK. The file is then uploaded to the site and validated. The message Loading will appear with a progress bar. Please wait for the file to upload, even if the progress bar doesn't move. This process should only take a few moments.

Once the file is uploaded, it is validated (checked against data requirements). An error message is given if the dataset doesn't validate successfully - for example, if any scores are not numeric, or marker identifiers contain disallowed characters. If you get a validation error message there will be information included about how to modify your input data in order to pass validation. For details on the acceptable file formats and how to lay out your data, please see data requirements - marker scores. Please contact us if you require guidance on validating your data.

Stage 2: Data annotation and processing

The upload dataset form, filled in with annotation details and settings

Figure 4: The add dataset form, showing dataset annotation and settings.

Once the file has been validated successfully, you will be asked to enter a few details and select processing options (figure 4). The options on the form are as follows:

Title: A title for the file. This is for your own reference. The maximum length is 48 characters.
Notes: (optional) Any notes you may wish to make about the dataset, again for your own reference, and can be edited later.
Scoring type: Categorical or continuous. A dataset is categorical when scores are in a small number of discrete categories (nine or fewer), e.g. using the Allred scoring system. Continuous datasets have scores which can take any value within a given range. Typically, the scoring type is automatically detected during the file validation stage and therefore can't be changed. The exception is small datasets where the scoring system can't be reliably identified. Please note that whole number scores should be used for categorical scoring. Currently, categorical data with more than nine scores are not supported - they will be treated as continuous. For such datasets, consider grouping score values together to produce nine or fewer different values, which will then be detected as categorical.
Batch correction: (datasets with continuous values and multiple batches only) This setting is used to indicate if batch correction is required. Batch correction is performed using the ComBat algorithm [Johnson et al. 2007 Biostatistics 8:118].
Replicate aggregation: (datasets with marker replicates only) This setting defines how scores for multiple replicates are combined to give a single score for a particular marker and patient. The available settings are mean and median. Mean is not available on datasets with categorical scoring. For more information on specifying replicates in datasets see data requirements - marker scores.

When the above has been completed, click the Save dataset button.

[Top]Accessing uploaded data

Figure 5: The dataset page showing processing status. Once the dataset has loaded, options for running and viewing analyses will appear.

Having clicked the Save dataset button, you will be redirected to a unique page for the dataset (figure 5). The dataset is added to a queue for processing, which typically takes just a few moments. Importantly, in order to return to the dataset, bookmark this page or make a note of the web address (URL). The URL contains a unique key which is needed to view the dataset.

Once the processing has completed, the dataset is ready and options for running analyses are made available.

If an error occurs during import, a message will be displayed that includes guidance on how to resolve the error. The help section on dataset error messages gives further information.

Figure 6: A freshly imported dataset page, showing that the dataset is ready to run analyses

For information on running analyses, see working with datasets.