Quantifying HiPSC-CM structural organization at scale with deep learning-enhanced SarcGraph

Saeed Mohammadzadeh and Emma Lejeune

PLOS Computational Biology, 2025, vol. 21, issue 10, 1-28

Abstract: In cardiac cells, structural organization is an important indicator of cell maturity and healthy function. Healthy and mature cardiomyocytes exhibit a highly organized structure, characterized by well-aligned almost crystalline morphology with densely packed and organized sarcomeres. Immature and/or diseased cardiomyocytes typically lack this highly organized structure. Critically, human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs) offer a valuable model for studying human cardiac cells in a controlled, patient-specific, and minimally invasive manner. However, these cells often exhibit a disorganized and difficult to quantify structure both in their immature form and as disease models. In this work, we extend the SarcGraph computational framework—designed specifically to assess the structural and functional behavior of hiPSC-CMs—to better accommodate the structural features of immature cells. There are two key enhancements: (1) incorporating a deep learning-based z-disc classifier, and (2) introducing a novel ensemble graph-scoring approach. These modification significantly reduced false positive sarcomere detections, particularly in immature cells, and improved the detection of longer myofibrils in mature samples. With this enhanced framework, we analyze an open-source dataset published by the Allen Institute for Cell Science, where, for the first time, we are able to extract key structural features from these data using information from each individually detected sarcomere. Not only are we able to use these structural features to predict expert scores, but we are also able to use these structural features to identify bias in expert scoring and offer an alternative unsupervised learning approach based on explainable clustering. These results demonstrate the efficacy of our modified SarcGraph algorithm in extracting biologically meaningful structural features, enabling a deeper understanding of hiPSC-CM structural integrity. By making our code and tools open-source, we aim to empower the broader cardiac research community and foster further development of computational tools for cardiac tissue analysis.Author summary: Heart disease remains a leading cause of morbidity and mortality worldwide. To better understand cardiac health and advance new therapies, researchers are increasingly turning to human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs). These cells provide a patient-specific platform for studying heart disease and developing engineered heart tissue. However, analyzing hiPSC-CMs remains challenging because their structural organization, especially in immature or diseased states, often appears disorganized under the microscope, making it difficult to extract meaningful quantitative insights. In this study, we present significant enhancements to our computational tool, SarcGraph, which is designed to detect and analyze sarcomeres, the fundamental units responsible for muscle contraction. Our improvements allow for more accurate detection of sarcomeres, even in structurally disorganized cells. We applied the updated tool to a large open-access dataset and demonstrated that it can not only replicate expert assessments of cell structure but also uncover inconsistencies in manual scoring and provide a robust automated alternative. By releasing our methods as open-source software, we aim to support the broader cardiac research community in studying hiPSC-CM structure and in developing more consistent, scalable approaches for assessing cellular growth, maturity, and disease-related changes.

Date: 2025
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1013436 (text/html)
https://journals.plos.org/ploscompbiol/article/fil ... 13436&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pcbi00:1013436

DOI: 10.1371/journal.pcbi.1013436

Access Statistics for this article

More articles in PLOS Computational Biology from Public Library of Science
Bibliographic data for series maintained by ploscompbiol ().