LabCAS ("Laboratory Catalog and Archive Services") is a
web-enabled environment that allows users to publish, share, search
and download a wide variety of biomedical datasets. This page contains
information for end-users of the LabCAS portal and services.
data can be published into LabCAS in two possible ways: by using the
LabCAS User Interface to upload and trigger ingestion of data; or by
uploading data directly to the LabCAS server. Before publishing, data
should be "curated" i.e. organized in a meaningful directory
structure, and provided with enough metadata to make them useful to
Data in LabCAS is organized according to the
following logical hierarchy:
broad sets of related data from the same study, the same analysis, or
the same project.
- Datasets: different sets
of related files within the same collection - for example, the
different patients of a clinical study, or the different labs
participating in a study.
- Files: all the
files in a given dataset - for example, all the images for a single
patient, or all the data supplied by a single
- Collection=RNA Sequencing,
Dataset=ERR164773, Files=ERR164773_1.fastq, ERR164773_2.fastq,
- Collection=University of Colorado Lung Images,
Dataset=UCHSC_1467, Files=22021 P06-32 B4 Bronchus Intermedium
Before publishing data into LabCAS, you
should organize them in a meaningful directory structure conforming to
the model above.
In LabCAS, metadata can be associated to
data at all 3 levels: collections, datasets, and files. When
publishing, it is recommended that the following metadata fields be
supplied to make the data searchable and more useful (mandatory fields
are underlined). When using the LabCAS UI to publish data, some
of these fields are automatically populated based on the user login,
or the value of other supplied fields.
CollectionName: short name
for high level data collection (example: MD Anderson Lung
CollectionDescription: a few sentences
describing the high level data collection (ex: Lung images for
clinical studies conducted during 2001-2010.)
uid=amos,dc=edrn,dc=jpl,dc=nasa,dc=gov - this is provided by the
Consortium (ex: MCL or EDRN)
Discipline: (ex: RNA Sequencing, Pathology,
LeadPI (ex: Chris
- Organ (ex: Lung, Pancreas,
- OrganId (ex: 3)
- Institution (ex:
Dartmouth, MD Anderson Cancer Research Center)
short name for this dataset (ex:
- DatasetDescription: a few sentences
describing this dataset (ex: Data for patient X)
asetURL (ex: http://someotherwebsite/patients?id=X)
- FileName: parsed from file
- FileLocation: parsed from file
- FileSize: parsed from file
other custom metadata can be supplied and stored as well, at all
levels of the data hierarchy.
Publishing via LabCAS UI
The LabCAS web
portal provides a web-enabled
workflow for uploading data to the server, and trigger publication to
the LabCAS archive. The user is guided through the process of
selecting the files from their own desktop, and populating the
required and recommended metadata fields. Because uploading data
through the browser is not very efficient, this process is recommended
for datasets that are not very large (up to a few GB in size).
Uploading Data to
the LabCAS Server
Very large datasets should be transferred to
the LabCAS server via some means other than a web browser. At this
time, LabCAS offers two ways to upload data directly: SFTP and WebDAV.
Both protocols are supported on a variety of platforms including
Linux, Mac OSX, and Windows. Before uploading data, please make sure
- Organize your data in a directory structure
corresponding to one collection and one or more datasets, as described
- For each dataset, provide as much metadata as possible
in a file colocated with the dataset, and named <dataset>.cfg
Then proceed through one of the two methods
In this section we
describe how to upload data using a Unix-like system (macOS, Linux,
FreeBSD, Solaris, etc.) and on Windows.
On Unix-like systems, do the
- Generate a public/private RSA key pair with the
following command, or use an existing pair if you have one:
- Send your public key (id_rsa.pub) to
the LabCAS team so they can enter it in the list of allowed user
- Simply use the SFTP client of your choice to connect to
the LabCAS server (with no username or password, since authentication
is provided by the key), navigate to your assigned root directory,
create new directories for the Collection and Dataset to upload, then
transfer the data (all files in the current local directory, including
the metadata file <dataset>.cfg):
On Windows systems, you
first generate a public/private key pair:
- Download and
install both PuTTY and WinSCP. Both are free, open-source
- Installing PuTTY makes a new program
available, PuTTYgen. Start this.
- At the bottom, select SSH-2
RSA and enter 2048 bits. (This should be the default.)
- Move the mouse around in the blank area
to generate random numbers until the progress bar is
- Click "Save public key" and save the key to
a file. Mail that file to JPL and we'll install it in the right place.
Alternatively, you can click in the Public key box, select ALL the
text, and paste that in an email. This key should be a single long
line of text that starts with "ssh-rsa AAAA…"
a passphrase (and confirm it) then press "Save private key".
Choose a filename (say "LabCAS") and save it to your
- Close PuTTYgen.
At this point,
you'll need to await confirmation from JPL that your public key has
been installed. Once you hear back, you can start uploading data with
- Open WinSCP.
- Press the
- On the left side, under
"SSH", click "Authentication".
"Authentication parameters", click the ellipsis … button
under "Private key file".
- Select the private key you
saved in the Documents folder in step 7 (above). Click
- Under "Session", select SFTP. For
Host name, enter zipper-vm.jpl.nasa.gov; for user name, enter
- You will be warned the first time you're
connecting to an unknown server. Press the "Copy Key"
button. Then press "Yes".
- Enter the passphrase for
the private key you created in step 7 (above). Check the box by
"Remember password for this session". Then press
- You now have a window on the left with your
local files and on the right. To send a file, select it on the left
and click "Upload". Or simply drag and drop. You'll be
prompted the first time to set up transfer settings. The defaults are
fine, so check "Do not show this dialog box again" and press
WebDav data upload
- Contact the LabCAS
team to obtain a WebDAV username and password, which will be stored in
the LabCAS LDAP database
- Use one of the supported clients on
your desktop to upload the data (see below).
After logging in with your credentials, drop your data in the location
- Mac OSX: macs come
with a pre-installed WebDAV client. In the upper bar menu, choose Go
> Connect to Server, and enter the URL above
- Linux: we
recommend using the cadaver client (see http://www.webdav.org/cadaver/):
Note that at this time uploading data through SFTP or
WebDAV will NOT automatically trigger data publishing (although it
will in the near future). Please contact the LabCAS team to let them
know new data is available, and to coordinate the publishing