Date of Award




Document Type

Master's Thesis

Degree Name

Master of Science (MS)


Department of Biological Sciences

Content Description

1 online resource (v, 63 pages) : color illustrations.

Dissertation/Thesis Chair

Alex Valm

Committee Members

ChangHwan Lee, Thomas Begley


Fluorescent, model, Multiplexed, optimization, simulation, unmixing, Fluorescence microscopy, Multiplexing

Subject Categories



Multiplexed fluorescent imaging (MFI) has transformed biological investigation by providing spatial context to labeled analytes, such as DNA, RNA, and proteins. Multiplexed imaging techniques can track multiple analytes at once and thus has the potential to provide an unprecedented view into the complex relationships between various molecules, cells, and tissues. However, most researchers rarely use more than 3 fluorophores at a time due to the complex technological and procedural requirements for highly multiplexed imaging. Currently, a major obstacle in the field is the inability to quantify uncertainty in fluorescent images, making it difficult to draw quantitative conclusions from the images. Further, uncertainty limits multiplexing to between 3-7 labels, a limitation exacerbated by the increasing complexity of optimizing microscopy settings with more labels. To solve this problem, we have built a microscope model capable of accurately simulating the mixture of multiple fluorophores. This model is capable of predicting linear unmixing quality in multiplexed experiments, which has not been done before. Further, have developed an algorithm for calculating uncertainty in existing multiplexed images, based on the capture settings and/or image metadata. This allows researchers to have an appropriate amount of trust or skepticism in their unmixed images. We have published the model on the web so researchers from around the globe working on widely available commercial instruments can optimize and quantify spectral experiments and images. This work will lay the foundation for designing a custom microscope capable of imaging multiple targets with the lowest uncertainty theoretically possible, resulting in new multiplexed capabilities far beyond our current technology.

Included in

Biology Commons