Vala Sciences

Download, install and launch CyteSeer from your Mac, Windows or Linux workstation. Most features are accessed from this CyteSeer main window.

All CyteSeer analyses start by breaking down each image into core biological compartment masks and then aggregates measurements as needed for experiments with many slides, wells, plates etc.

1) First, CyteSeer identifies all of the nuclei available from the nuclear images. A nuclear mask for each cell is established where the mask contains all of the pixels locations automatically identified as nuclear for a given cell.

2) Next, CyteSeer analyzes a second image – a lipid droplet image in this example – such that the lipid droplets are assigned to the lipid droplet mask. A rich set of data parameters are then calculated on a “per cell basis”.

3) Third, CyteSeer tries to define the cell boundaries. This varies from assay to assay. It can be done by cell border, cell membrane or – as in this case – by roughly estimating which droplets belong to which cell.

CyteSeer organizes data on a cell by cell basis. Each cell is a unique record in CyteSeer's database with a rich set of measurements for size, shape, intensity and colocalization based on whichever algorithm is chosen. CyteSeer can arbitrarily filter and report subtle changes in cell behavior with a rich set of gating tools.

CyteSeer is installed with example images, image masks and numeric data. Use the Image Viewer compare the original images with the image masks that CyteSeer's algorithm create. Adjust the masks by changing nuclear size and channel's sensitivity on the main CyteSeer window and clicking the "Run" button.

The 'Set Colors" dialog controls a central powerful aspect of CyteSeer.

The original images from your microscope are visually combined with the masks generated by CyteSeer. The masks are the result of CyteSeer's automated cell analysis engine and all the subsequent numeric data are derived from these masks. If you want to count how many lipid droplets (green) there are on a cell by cell basis, you'll want to visually check how well the masks match the original images. In this example, we're looking at a nuclear stain in blue and a lipid droplet stain in green. Masks outlining the edge of the nucleus and lipid droplets are shown in red and green respectively. The yellow lines define the cell regions and show the unique cell ID for each cell that CyteSeer finds.

No image processing is perfect over every cell in every image, but this image viewer allows you to get a feel for how well the current analysis settings are working.

In the image above, an overlay of yellow and red outlines of the automatically generated masks of the nuclei and lipids are shown.

Due to variable staining and imaging conditions, an algorithm's sensitivity may need adjusting.

Adjust sensitivity and nuclear size first: If the masks do not meet expectations, then the images need to be reprocessed with new sensitivity and/or nuclear size settings. This is repeated heuristically until the you are either satisfied with how well the algorithm is generating a representative mask or decide that a more sophisticated algorithm is required.

Develop new algorithms yourself or with Vala's help: CyteSeer has an integrated graphically driven algorithm development tool available for advanced users. Vala also actively works with many groups to develop and deploy new algorithms.

As imaging conditions vary wildly from sample to sample and system to system, it often makes sense to rerun the samples with a different sensitivity and/or size value on the main CyteSeer window.

If CyteSeer creates masks that under-represent certain parts of the cell, the sensitivity is too low. If the mask is over-represented the sensitivity is too large. In routine use, CyteSeer will likely work best with a sensitivities of 25% to 400%.

The size values most often refer to nuclear size. If CyteSeer incorrectly identifies a large proportion of single nuclei into multiple nuclei, the size value is too low. If two nearby nuclei are counted as one, the size value is too large. In routine use, CyteSeer will likely work best with a nuclear sizes from 3 to 30.

The data viewer can view cell-by-cell results within an individual well or across regions of one or more plates.

The default example data within CyteSeer demonstrates a increase in the average per cell count of lipid droplets from row B to row G.

1) Select "Row" from PlateMap Quantity to tell CyteSeer that you'd like to see the row by row response
2) Type in "count" in the search tool to find the measurement of interest
3) Click on the measurement to generate a bar graph. Save all graphs via the right mouse click

The "Analyze Statistics" button of the Data Viewer opens up a powerful analysis tool.

Note that selecting "Average Cells" followed by "Well" for PlateMap Quantity will display a the well, cell count and mean value of the selected measurement – the "Average of Lipid Droplet Count" per cell in this case.

Use the "Well Data" tab to explore data specific to an individual well. The gating and cell gallery tools are powerful cell-by-cell "Well Data" tools.

Each cell has a unique cell ID matched in both the Image and Data Viewer.

This has been a brief introduction to CyteSeer membrane segmantation algorithm. As you learn more, keep in mind that CyteSeer has powerful gating, cell gallery and integrated custom algorithm creation tools as well. CyteSeer will work with most images from most digital fluorescent or bright-field microscopes.

Please explore our Vala applications and CyteSeer tutorials section of the website for more detail on what else can be done. If what you're looking for isn't there, please don't hesitate to contact us with more detailed questions directly or via our public forum.

Vala Sciences is a cell biology company offering software, kits, services and custom development for analyzing a wide variety of cell types and conditions from adipocytes & stem cells to primary & well established cell lines.

For further information on products and ordering information please call or email:
Vala Sciences Inc.:
Sales and technical information:
858 461-6861
info@valasciences.com
http://valasciences.com

CyteSeer is a registered trademark of Vala Sciences Inc. All rights reserved 2005-2009.