TACITuS (Transcriptomic Data Collector, Integrator, and Selector) is a web portal that simplifies the process of collection, pre-processing, selection, and integration of transcriptomics data. Therefore, users can collect data from major sources, such as NCBI GEO or ArrayExpress, and integrate them with their own data into a standardized format, facilitating subsequent analyses. Our software is freely available and distributed through a GPL v3 licence. TACITuS implements five major functionalities: (i) data import, (ii) data selection, (iii) identifier mapping, (iv) data integration, and (v) Galaxy Export, which are shown later.

Importing data

Data can be imported through the Dataset panel. There the user will be able to select a data source (NCBI GEO, ArrayExpress, or custom). When a public source is selected (NCBI GEO or ArrayExpress) the accession number of the dataset he wishes to import has to be provided. For a custom dataset the user must also provide a name, a metadata file and the expression file. Metadata file should be uploaded as a tab-separated value file where each row corresponds to a sample and each column is a covariate. Headers for each covariate should be provided. Sample identifiers have to be positioned in the first column of the file. Expression file should be formatted as a tab-separated value file where each row corresponds to a probe and each column to a sample. Probe identifiers have to be positioned in the first column of the file, and Sample identifiers need to be positioned in the first row.

As soon as all required data are provided the user can decide whether the imported dataset should be public or not and submit the request. Then, a record is added in our processing queue and a progress log is displayed in the Jobs panel.

Selecting data

For each dataset listed in the Dataset panel, the user can press Make Selection to start the selection process. A panel will appear where the user is required to specify the selection name and which samples should be included. Samples can be selected through the metadata table where all covariates for each sample are displayed. On the bottom of the table a search field for each covariate is provided. Samples can be selected by clicking on a row in the table. Bulk selection and deselection of samples is also provied. Select All/Deselect All allows selection/deselection of all samples in all pages of the displayed table. Select Shown/Deselect Shown allows selection/deselection of all samples in the displayed page of the table.

By pressing Submit selection the user submits his request. Once sent, a record is added in our processing queue and a progress log is displayed in the Jobs panel. Results can be downloaded in TSV or CSV format.

Mapping datasets

To facilitate the data integration process with other platforms, TACITuS supports mapping from probe identifiers to standardized one (e.g Entrez Id or ENSEMBL Id) for each selection. The functionality is activated through the Selections panel by clicking on Map Identifiers. Once the Map Identifiers panel is open, the user will be able to select a platform (the field is automatically detected for NCBI GEO dataset) and a destination identifier. Destination identifiers depends on the specific experimental platform. By pressing Submit, the user submits his request. Once sent, a record is added in our processing queue and a progress log is displayed in the Jobs panel. At the end, the user may download the mapped data in TSV or CSV format.

Integrating datasets

TACITuS implements several integration procedures which combines two or more selections into a single dataset. The system exploits the following techniques:

• Sims et al. 2008 transforms each dataset through mean-centering applying a technique like z-score normalization.
• COMBAT exploits an Empirical Bayesian model to estimate the mean and variance of each gene in each dataset, correcting data for batch effects.
• Gene Standardization is the simplest mathematical transformation to make datasets comparable. For each gene, the expression value is corrected by subtracting the mean and dividing by the standard deviation.
• XPN finds blocks of genes and samples showing similar expression patterns. The average of these blocks will be used to shift and scale the data.

The Integration procedure is activated by clicking on Request integration in the Integrator panel. The user can specify which datasets to integrate, the combination method, and a supplementary identifier mapping. Integration algorithms can also be disabled if the user needs to combine expression matrices without altering the values. Once the request has been completed, a job is added to our jobs queue, and processed as soon as the resources are available. The results can be downloaded from the Integrator panel both in TSV or CSV format.

Uploading to Galaxy

TACITuS is equipped with a module for uploading data and metadata to the Galaxy platform. First, the user should register to a galaxy server. Then, he provides the credential in the Galaxy Account panel (name, host name, port, and API key). As soon as these data are entered the upload module will be enabled.

To upload a dataset, the user has to click on Upload to Galaxy present in each panel of the website (Selections, Mapped Selections, Integrations). A page will open and therefore the user can select a galaxy server. By clicking on Upload a background upload job will start, and the user will be notified as soon as the process is completed.