Importing Data
Using the Data Browser in the KBase Narrative Interface.
Last updated
Using the Data Browser in the KBase Narrative Interface.
Last updated
The Import tab lets you drag & drop data files from your computer into the Staging Area to upload data into your Narrative. Visit this page of the Narrative User Guide for general instructions on how to use the Staging Area.
The first step in uploading your data is to locate the Data Panel along the left side of the Narrative Interface window and click the “Add Data” button, the circular “+” icon, or the arrow at the upper right of the panel to access the slide-out Data Browser. The Data Browser has several tabs that allow you to access data within KBase and the Import tab for uploading your own data.
To close the Import panel and return to the Narrative Interface, simply click the “Close” button on the bottom right of the import panel.
If you have a small screen, you might not be able to see that button. Another way to close the Data Browser is to click the arrow icon on the top right of your Data Panel (the same one that opens the Data Browser slide-out).
The import type selection dropdown can detect the data type of the file based on its extension where possible. Some types will not automatically select, such as .fasta or .fastq. In these cases more than one importer can use the FASTA file extension. However, suggested importers will be listed at the top of the drop down under “Suggested Types.”
Learn how to import multiple files under the Bulk Import Guide
For example, files with the extension FASTQ can import as a single interleaved file, as a pair of forward and reverse non-interleaved files to create a PairedEndLibrary, or as a single non-interleaved file to create a SingleEndLibrary.
To import a file, make sure you have an import type selected and the check box is active, then click "Import Selected."
The drag & drop from your local computer works for many files. There is a size limit around 2GB, which is dependent on your computer and browser. For larger files, use the Globus Online transfer or a direct upload with a URL.
Any data you upload to KBase is private unless you choose to share it. You can share any of your Narratives (including their associated data) with one or more specific users, or make it publicly available to all KBase users. Please see the Share Narratives page for more information and how to share data and Narratives. KBase policies on data are described on the Terms and Conditions and Data Policy.
The next sections of this guide describe the specific steps involved in uploading the currently supported data types, and show examples for each type.
If you are a JGI user, you can transfer public genome reads and assemblies (including your private data and annotated genomes) from JGI to your KBase account—see this page for instructions.
To import multiple files, ensure all the files you want to import have an import type selected and the check box is active, then click "Import Selected."
Bulk import currently supports one parameter set per data type. If you need to upload files of a given data type with different parameters, perform a separate set of bulk imports for each parameter set.
You’ll see a new import cell created for bulk imports. This cell contains a tab for each of the data types you selected from the Staging Area, which you can view by clicking on the available types within the Data type column on the left. This cell is a bulk wrapper for existing import apps, so you can fill out the parameters the same as you would for a single import.
Once you have filled out all file paths and parameters, run the import. While the import is running you can see all the logs and status details in the Job Status tab, formatted for the bulk import.
Expand each of the child jobs by clicking on the corresponding line to view details, then expand further to see individual logs. When the bulk import cell is set to run, each child job for the individual data object will have a status, including Actions to cancel or retry individual jobs.
Alternatively, queued and running jobs can be cancelled using the bulk action dropdown. Or the retry action can be used for multiple import jobs within the bulk import cell.
When the jobs are completed successfully, you’ll be able to see all the information on the data object, repackaged for viewability in bulk. The first new feature is a table of all successful jobs. Clicking on object names adds a viewer widget to the Narrative. All the reports are available under the Reports tree. The full list is collapsed by default to conserve memory. The list can be expanded to show each report, and each report can be viewed in a separate window.
While this will be applicable to all data types and file extensions in the future, bulk import currently supports the following:
Assembly - FASTA
SRA Reads
FASTQ Reads Interleaved - FASTQ Interleaved reads
FASTQ Reads Noninterleaved - FASTQ paired-end reads or FASTQ single-end reads library
GenBank Genome - GenBank
GFF Metagenome - FASTA and GFF3
When selected files are not supported data types for Bulk Import within the Staging Area, separate, single import app cells will open in the Narrative for each type and file.
This is a new feature in KBase, and the current release should be considered a beta version with future development still to come. Some of the most notable limitations are listed below. This is not a comprehensive list, but does contain the known bugs, issues, and limitations that are the highest priority for future releases. Please report any new bugs you find to the Help Board.
