General:
What is the YRC Public Image Repository?
The YRC Public Image Repository (PIR) is a large database of fluorescence microscopy images and
their associated metadata/experimental parameters. It includes a web interface, built for searching and viewing
the image data within the database. It is meant to be a resource for computational biologists
interested in problems of image analysis, and the automated analysis and annotation of microscopy
images. Biologists may also use the YRC PIR to obtain images depicting proteins of interest or
images depicting proteins that are annotated with GO terms of interest.
The YRC PIR is meant to be purely an image database, and includes annotations of images that
describe parameters from their respective microscopy experiments--such as tagged proteins, fluorescent tags used, emission
and excitation wavelengths, shutter speed and so on. The YRC PIR is not meant to be an extensive protein annotation
database, and does not include metadata about proteins beyond their Gene Ontology annotations. For extensive
protein annotation information, please visit the YRC PDR.
What is the YRC?
The Yeast Resource Center (YRC) is a multi-disciplinary NCRR Biomedical Technology Resource Center
focused on exploiting the budding yeast
Saccharomyces cerevisiae as a model organism to develop tools
and technologies to characterize proteins and proteomes, which may then be applied to other systems.
Principle technologies employed by the YRC include protein mass spectrometry, protein structure prediction
and design, fluorescence microscopy (including fluorescence energy transfer), yeast two-hybrid, and
development of novel computational biology algorithms. In addition to core technology development
projects, the YRC participates in hundreds of proteomics collaborations each year with researchers
around the world.
To learn more about the YRC follow this link.
Who funds the YRC PIR?
How do I cite the YRC PIR?
Please use this citation: Riffle M, Davis TN (2010) The Yeast Resource Center Public Image Repository: A large database of fluorescence microscopy images. BMC Bioinformatics. 2010 May 19;11(1):263.
Who built the YRC PIR?
The YRC PIR was built by Michael Riffle under the direction of Trisha Davis at the University of Washington in Seattle.
For more information see the
About Us section.
How do I download data?
The entire data set may be downloaded as uncompressed, unscaled 16-bit images from
the
download page. Alternatively,
the results matching the current search may be downloaded at any time in the YRC PIR web interface
by clicking "Save as Bulk Download" button. More information about downloading data may be found in
our
User Guide.
Can I add my data to the YRC PIR?
Not at this time, though this feature is planned. In the mean time, you may upload small sets of
high quality microscopy data for any protein to the
YRC PDR.
To upload data, search for and navigate to the protein for which you have images and click the
"Upload Your Own Microscopy Data" link in the Microscopy section.
About experiment metadata:
Z-sections
The number of focal planes along the Z-dimension in an experiment.
Z plane depth
The depth of field for each pixel. This is a measure of the distance between the furthest and nearest objects simultaneously in focus in an image.
Time series
The number of times a single color channel (distinct emission and excitation wavelength combination) was imaged at the same X, Y, Z stage position in an experiment.
Panels
The number of distinct X, Y stage coordinates in an experiment.
Objective
The magnification of the objective used in an experiment.
Binning
Binning is the process of combining sets of adjacent pixels into a single pixel to reduce noise and improve signal to noise ratio.
A binning of 2x2 combines a 2x2 block of 4 pixels from the camera into a single pixel in the final image. More information may be
found
here.
Image size
The X and Y dimensions of images in an experiment, in pixels.
Pixel size
The physical dimensions represented by a single pixel in images from an experiment.
Species
The species to which the tagged proteins in an experiment belong.
Strain
A string identifier for a specific genetic strains of yeast that may contain antibiotic resistances, fluorescently-tagged proteins of interest or other markers. You may download a mapping
of the strain names to genotype definitions from our
download page.
Channels
A listing (or count) of the unique combinations of excitation and emission wavelengths in an experiment. These typically include a DIC channel, one or more color channels (in which particular
fluorescent proteins are excited and observed) and a FRET channel (in the case of a FRET experiment).
About image metadata:
Filter Channel
The name of a particular combination of excitation and emission wavelengths. Serves as a human-readable guide for which fluorescent protein tags should be visible in a particular image.
Visible Protein
The tagged protein that should be visible in a particular image, given its fluorescent tag and the emission wavelength of the image.
Donor and Acceptor Protein
Only applicable to the FRET channel in FRET experiments. The donor protein is the protein tagged with the fluorescent protein excited at the excitation wavelength. The acceptor protein's fluorescent
tag accepts energy from the donor protein's tag and is imaged using the emission wavelength in the FRET channel.
Excitation and Emission λ
The excitation wavelength is the wavelength of light used to excite the fluorescent protein tag in an image. The emission wavelength is the wavelength used to detect the resulting fluorescence.
Time Point
The time, in seconds, a particular image was taken after the start of an experiment.
X, Y and Z Coord
The stage coordinates for a particular image. Changes in the X and Y coordinates denote lateral movements and changes in the Z coordinates denote axial movements.
Photosensor
The measured brightness of the lamp used to excite the fluorescent tag.
Exposure
The time, in seconds, the shutter was open when taking an image.
NDFilter
Describes the level of neutral density filter applied to the image.
About the Data:
Where did the images come from?
How were the images acquired?
What format are the images in?
The images presented within the pages of the web site are PNG images. They are produced on-the-fly by
converting underlying TIFF data in the database to PNG images so that they may be displayed in a web browser.
All images downloaded from the YRC PIR will be uncompressed (raw)
16-bit OME TIFF images.
What is OME TIFF?
OME TIFF images are TIFF images that contain XML conforming to the OME XML specification in the TIFF header. More information
regarding OME TIFF and OME XML may be found at
http://www.ome-xml.org/. OME
TIFFs will be treated exactly as "normal" TIFF images by image analysis software. Programs that support the OME XML
specification, such as ImageJ (with installed modules) will be able to make use of the additional information in the TIFF header.
How are the images scaled by the web site? Why?
By default, pixel values in images are scaled for viewing in the web site. This is done because typical
images generated by flourescence miscroscopy are very dim, and would appear black or nearly black to the
naked eye.
Pixel values, or intensity values, are rescaled for viewing using the following algorithm: The minimum (min)
and maximum (max) intensity for the image are found. The intensity for each pixel is then set as:
( ( pixel intensity - min ) / ( max - min ) ) * 65535
(note: this formula presumes 16-bit greyscale images.)
This scaling may be toggled on and off when viewing experiments by clicking "Turn Scaling OFF/ON." Note: Image data
downloaded from the YRC PIR is always unscaled.
What are the image statistics? How are they calculated?
The image statistics optionally display the minimum intensity, maximum intensity, mean intensity and a histogram depicting
the distribution of intensities in a microscopy image. The values are determined by reading the intensity value for every
pixel in the image.
The histogram depicts the frequencies of all possible intensity values in the image. The scale of the x-axis is 0 to
65535 (the possible intensity values in a 16-bit greyscale image). The scale of the y-axis is the ratio of the number
of pixels with a given intensity to the number of pixels for the intensity with the most pixels. The y-axis essentially
depicts the frequency of a given intensity relative to the most frequent intensity.
NOTE: If scaling is turned ON, the statistics and histogram are derived from the scaled image.
Image statistics may be toggled on and off when viewing experiments by clicking "SHOW/HIDE image statistics."
What are the scale bars? How are they generated?
The scale bars show the space in the image that represents 10 microns. Their pixel length is determined by dividing 10
by the length dimension of the
pixel size for the experiment. They are
generated on-the-fly and applied directly to the image shown in the web site. The scale bars may be toggled on and off
by clicking "SHOW/HIDE scale bars". Images downloaded using the
download page
or by defining bulk downloads will not contain scale bars.
Why were the included proteins chosen?
The included list of proteins are the complete set of proteins depicted in the images available from the Davis laboratory
for inclusion in the YRC PIR, taken over the course of several years. This particular list of proteins reflect
those studied by this research group during that time. As more data is added to the YRC PIR, the list of covered
proteins will continue to grow.
Where do the GO terms come from?
Gene Ontology (GO) terms associated with images in the YRC PIR are the GO terms associated with the tagged
proteins in respective experiments, as taken from the Gene Ontology database. They are not the product
of human curation of each image, and are automatically assigned based on the known annotation of depicted
proteins.
About the Web Site:
What software is used to run the site?
The YRC PIR web application runs on Linux, utilizing the Apache Struts servlet/JSP framework,
the Apache Tomcat servlet container and the Apache httpd web server. All data presented
in the YRC PIR is stored using the MySQL relational database management system.
What programming language was used?
The YRC PIR was entirely developed utilizing the Java programing language. It makes use of the Java
Advanced Imaging API.
