Print  Close

HistoJS: Web-Based Analytical tool for Advancing Spatial Biology

Poster No:

1948 

Submission Type:

Abstract Submission 

Authors:

Mohamed Masoud1, Sergey Plis1

Institutions:

1Georgia State University, Atlanta, GA

First Author:

Mohamed Masoud  
Georgia State University
Atlanta, GA

Co-Author:

Sergey Plis  
Georgia State University
Atlanta, GA

Introduction:

Advances in multiplexed imaging technologies enable us to capture single-cell proteomics and transcriptomics data in unprecedented detail and with high spatial resolution at the single-cell level. This large volume of image data presents a challenge in accurately isolating and quantifying distinct cell types to gain a deeper understanding of brain complexity, neurological disorders, potential biomarkers, and druggable targets for drug development. Therefore, there is an incremental demand and challenges to develop and validate cutting-edge quantitative image analysis tools for diagnosis, prognosis, and therapy response prediction and assessment in neurological and oncological diseases. HistoJS is a newly developed web-based tool that aims to overcome the challenges of utilizing highly-multiplexed immunofluorescence (HMIF) images for Spatial biology research. It provides open-source and extensible tools for analyzing spatial-molecular patterns, enables a deep view into the single-cell spatial relationship, along with machine learning algorithms in an interactive graphical interface that is easy-to-use for non-experts.

Methods:

Immunofluorescence images, representing protein type intensities in tissue, are managed and hosted using the Digital Slide Archive (DSA), an open-source platform [1]. DSA, comprising components such as MongoDB, Girder worker, and Girder, serves as a reliable web-based platform for storing and organizing large immunofluorescence image datasets. Installable locally or on the cloud, DSA facilitates user access controls and programmatic data management. HistoJS, utilizing DSA as a backend, enables users to manage and store images along with analysis results based on preferences. OpenSeadragon [2], an open-source viewer, is employed for visualizing Highly-multiplexed immunofluorescence (HMIF) [3-4] images in HistoJS, offering features like panning, zooming, and overlays. HistoJS supports customized protein channel stacking, enabling high-resolution and full-scale composite rendering (Fig. 1). Analytically, HistoJS deploys Flask APIs to encompass cell segmentation [5], phenotyping, classification, correlations, spatial analysis, and cell type quantification for understanding cell interactions. Real-time challenges are addressed with efficient techniques, including rapid cell boundary extraction and morphological feature analysis (e.g., solidity, eccentricity, orientation). Cell neighbor detection is facilitated through spatial plotting and Delaunay graph computation.
Supporting Image: Screenshot2023-12-01at23-37-06CopyofHistoJSNextGenerationWeb-BasedNeuroimagingApplicationdocx.png
 

Results:

HistoJS source code is publicly accessible with detailed step-by-step documentation in the GitHub repository https://github.com/Mmasoud1/HistoJS. Multiple DSA online servers can be accessed from HistoJS such as https://styx.neurology.emory.edu/girder/# to load HMIF data samples and test the tool performance (Fig. 2). Researchers could visualize and analyze the expression patterns of key biomarkers associated with neurodegenerative disorders. The platform's interactive features facilitated the identification of disease-specific signatures, providing valuable insights into the molecular basis of brain diseases.
Supporting Image: Fig2.png
 

Conclusions:

HistoJS is among the first web-based tools to visualize and analyze highly-multiplexed immunofluorescence (HMIF) image data while minimizing user inputs. For many researchers, setting up such analytical tasks is a technological barrier, and providing them through an interactive web-based pipeline will help biomedical users and related groups understand the progression of biological diseases and clinical outcomes in an easy-to-use and interactive graphical user interface. Future developments will focus on expanding the tool's functionalities and enhancing its accessibility to further accelerate advancements in neuroscientific research.

Genetics:

Transcriptomics

Modeling and Analysis Methods:

Methods Development 1

Novel Imaging Acquisition Methods:

Imaging Methods Other 2

Keywords:

Acquisition
Data analysis
Data Organization
Design and Analysis
Development
Workflows

1|2Indicates the priority used for review

Provide references using author date format

[1] Gutman, D.A., Khalilia, M., Lee, S., Nalisnik, M., Mullen, Z., Beezley, J., Chittajallu, D.R., Manthey, D. and Cooper, L.A., 2017. The digital slide archive: A software platform for management, integration, and analysis of histology for cancer research. Cancer research, 77(21), pp.e75-e78.
[2] OpenSeadragon dev. Team. (2012). OpenSeadragon. In GitHub repository. GitHub. https://github.com/openseadragon/openseadragon
[3] Lin, J. R., Fallahi-Sichani, M., & Sorger, P. K. (2015). Highly multiplexed imaging of single cells using a high-throughput cyclic immunofluorescence method. Nature communications, 6(1), 8390.
[4] Lin, J.R., Izar, B., Wang, S., Yapp, C., Mei, S., Shah, P.M., Santagata, S. and Sorger, P.K., 2018. Highly multiplexed immunofluorescence imaging of human tissues and tumors using t-CyCIF and conventional optical microscopes. elife, 7.
[5] Schmidt, U., Weigert, M., Broaddus, C., & Myers, G. (2018). Cell detection with star-convex polygons. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2018: 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part II 11 (pp. 265-273). Springer International Publishing.