Gastruloids are engineered from human pluripotent stem cells carefully cultured into circular microcolonies on specialized substrates. By adding small signaling molecules, these colonies undergo cellular differentiation mimicking gastrulation, a pivotal process in which the embryo begins organizing distinct cell lineages destined to form the body’s tissues and organs. Despite their promise as ethical alternatives to studying early embryos, extensive research has been hampered by inefficient manual methods that struggle to isolate and analyze these microscopic structures individually.

The team’s innovation addresses this bottleneck with an integrated sorting platform that combines high-resolution microscopy, imaging technology, and robotic micromanipulation. Central to this system are the “microrafts” — tiny, detachable supports on which gastruloids grow. Using a custom-built stage controlled by bespoke software, the system autonomously scans vast arrays of microrafts, identifies gastruloid characteristics, and physically sorts them based on preset criteria. This mechanized precision accelerates experiments that would otherwise be tedious and prone to human error.

Ian Jan, the lead author, likens the sorting mechanism to the claw machine games popular in arcade settings but designed to delicately “grab” microrafts harboring these miniature embryos. By automating this process, researchers can now harvest virtually hundreds of consistently sized gastruloids simultaneously and subject them to individualized analyses, thereby facilitating a high-throughput approach to developmental studies unprecedented in this domain.

What makes gastruloid models particularly valuable is their capacity to reveal early developmental abnormalities that mirror human diseases. Prior studies utilizing these models have illuminated how conditions like Huntington’s disease manifest genetically during the very earliest embryonic stages, even before traditional clinical symptoms emerge. Moreover, these platforms enable scientists to explore the phenomenon of aneuploidy—where cells contain abnormal numbers of chromosomes—and to observe the embryo’s inherent mechanisms for self-correction, shedding light on the remarkable resilience of early development.

The newly developed sorting system is poised to deepen these insights by making it feasible to isolate unique gastruloids that exhibit subtle phenotypic differences. This capability is essential given that heterogeneity among gastruloids reflects the complexity of authentic human embryogenesis, where individual cells and clusters follow diverse developmental trajectories. Recognizing and probing this diversity offers the potential to untangle how various genetic and epigenetic factors influence embryonic outcomes.

From a technical perspective, the sorting apparatus integrates a high-sensitivity camera with angular precision optics, coordinated by an advanced software pipeline that employs image processing algorithms to rapidly identify target gastruloids. Once selected, microrafts are released and collected in specific chambers, enabling downstream molecular or functional assays. This efficient, non-destructive method preserves sample integrity and opens doors to subsequent live-cell imaging or genetic sequencing examinations.

Looking ahead, the research team is ambitiously working to incorporate neural networks and machine learning models into the image analysis process. This artificial intelligence integration promises to enhance sorting accuracy and recognize complex patterns that elude conventional algorithms. Ultimately, such computational augmentation will enable real-time, adaptive screening of gastruloid phenotypes across vast experimental datasets, bringing unparalleled speed and sophistication to studying human developmental biology.

The significance of this technology extends beyond fundamental research. By improving our capability to screen large gastruloid arrays, the platform holds implications for drug discovery, toxicology assessments, and personalized medicine initiatives. Pharmaceutical companies might one day use automated gastruloid sorting to investigate how candidate compounds affect early human tissue organization, substantially reducing reliance on animal models and expediting regulatory approval workflows.

This advance exemplifies the kind of interdisciplinary convergence where engineering principles meet biological complexity. The marriage of microscale robotics with stem-cell science delivers a powerful toolset to tackle long-standing challenges in embryology, empowering researchers to interrogate developmental pathways with unmatched throughput and precision. Through this lens, the intricate dance of cells during gastrulation can now be observed, manipulated, and understood in transformative new ways.

Overall, this large-scale sorting system marks a pivotal step toward decoding the mysteries of human development. By enabling extensive, automated analysis of gastruloid variations, it shines a spotlight on the intrinsic heterogeneity fundamental to growth and differentiation. As the platform continues to evolve with AI-driven enhancements, it heralds an exciting era where the earliest chapters of life can be studied systematically, ethically, and at scale, unlocking insights that may ultimately improve human health across generations.

Subject of Research: Development and automated sorting of gastruloids to study early human embryonic development and detect aberrant developmental phenotypes.

Article Title: Development of large-scale gastruloid array to identify aberrant developmental phenotypes

News Publication Date: June 10, 2025

Web References: https://doi.org/10.1063/5.0269550

References: Jan, I., Cearlock, A., Yang, M., & Allbritton, N. L. (2025). Development of large-scale gastruloid array to identify aberrant developmental phenotypes. APL Bioengineering. https://doi.org/10.1063/5.0269550

Image Credits: Jan et al.

Keywords: Embryos, Embryology, Life sciences, Developmental biology, Physics

One of the overarching themes of City of Hope’s presentations is the integration of artificial intelligence (AI) and multiomics to decode the complex biology of tumors. Professor David W. Craig, an authority in integrative translational sciences, chairs the final plenary session titled “Opportunities in Predictive Oncology.” This session will explore emerging computational and biological strategies that leverage multi-level tumor data to refine precision oncology. Dr. Craig’s work particularly focuses on melding diverse data types—including genomic, proteomic, and spatial information—to dissect tumor heterogeneity and treatment resistance, fundamental barriers in effective cancer therapy.

Dr. Craig also helms an educational session dedicated to AI and data science, spotlighting how multi-scale, multi-modal integration enhances understanding of cancer’s genetic diversity. Highlighting methods such as spatial transcriptomics and single-cell genomics, this session illustrates how dissecting the spatial architecture within tumors reveals subclonal variations influencing tumor progression and therapeutic response. Graduate researcher Nina Song from Dr. Craig’s laboratory will present novel findings demonstrating the power of AI to fuse digital pathology with genomic data, offering unprecedented insights into aggressive cancers like glioblastoma, triple-negative breast cancer, and colorectal cancer.

Natural killer (NK) cells represent another focal point of City of Hope’s scientific agenda at AACR 2025. Michael A. Caligiuri, M.D., former president of City of Hope National Medical Center, chairs critical educational sessions on the biology and clinical translation of NK cells. These innate immune lymphocytes have emerged as potent anti-cancer effectors, capable of recognizing and eradicating transformed cells without prior sensitization. Dr. Caligiuri’s presentations will delve into molecular mechanisms regulating NK cell function and therapeutic strategies leveraging NK cells as immunotherapy agents, reflecting City of Hope’s leadership in harnessing innate immunity to combat cancer.

In addition to immunology, City of Hope scientists are pioneering research in precision medicine for underserved populations, an imperative often overlooked in cancer research. Postdoctoral scientist Francisco Carranza will unveil multi-omics analyses dissecting the MYC oncogene and WNT signaling pathway alterations in early-onset colorectal cancer among Hispanic/Latino patients. By integrating genomic and spatial transcriptomics technologies, this research elucidates the molecular underpinnings of cancer disparities and guides development of tailored diagnostics and treatments informed by ethnic diversity.

Precision artificial intelligence tools for clinical oncogenomics represent a further area of innovation. Assistant Professor Enrique Velazquez Villarreal and collaborators have developed PM-AI Agent, a conversational AI system designed to integrate extensive clinical, genomic, and social determinants of health data. This tool aims to facilitate equitable precision oncology by accounting for population-specific variables and social factors, providing clinicians with actionable insights that transcend traditional data silos. Such integrative approaches promise to reduce disparities and optimize therapeutic decision-making in complex cancer cases.

City of Hope’s clinical trial portfolio also features prominently at AACR 2025, highlighting advances in antibody-drug conjugates (ADCs) and immune checkpoint inhibitors. Hope Rugo, M.D., newly appointed director of the Women’s Cancers Program, will present on managing toxicities associated with emerging ADCs, which combine targeted antibodies with potent cytotoxins to selectively eliminate cancer cells. Her expertise also extends to discussions on biologics and T-cell engagers, signaling ongoing efforts to refine immune-based therapies in breast and other cancers.

Among other high-impact clinical presentations, Aditya Shreenivas, M.D., M.S., will report phase 3 trial results for Penpulimab, a humanized anti-PD-1 monoclonal antibody evaluated as first-line treatment for recurrent or metastatic nasopharyngeal carcinoma. These findings could redefine therapeutic options for this aggressive malignancy by improving survival and tolerability in diverse patient populations, reflecting City of Hope’s commitment to global oncology.

Research connecting fundamental biology to therapeutic resistance mechanisms will be illustrated by Kimya Karimi, a postdoctoral scholar investigating ways to overcome cell cycle inhibitor resistance in estrogen receptor-positive (ER+) breast cancer. By combining epigenetic and molecular analyses, this work aims to restore endocrine therapy efficacy, addressing a significant clinical challenge in breast cancer management.

The conference also spotlights the vital role of community engagement in translating scientific discoveries into health policy and patient outcomes. Kimlin Tam Ashing, Ph.D., will elaborate on frameworks for fostering community alliances and partnerships that promote equitable cancer care delivery. This integration of social science with biomedicine embodies City of Hope’s holistic vision of research impacting patients beyond the laboratory.

Highlighted poster sessions further reveal City of Hope’s versatile expertise. Senior research associate Jing Qian will present spatial transcriptomic data unmasking differences in tumor and immune microenvironments among high-grade serous ovarian cancers, providing insights into variable responses to checkpoint blockade immunotherapies. Similarly, hematology fellow Peter Zang will explore spatial proteomic distinctions in metastatic prostate cancer across ethnicities, informing biomarker development and personalized treatment strategies.

Lastly, cutting-edge computational approaches to cancer prognosis are represented by a team including postdoctoral fellow Sydney Grant and assistant professor Aritro Nath. Their application of survival-based variational autoencoders to integrate multimodal data advances predictive modeling of recurrence-free survival in breast cancer patients, potentially guiding individualized risk assessment and therapeutic planning.

City of Hope’s robust presence at AACR Annual Meeting 2025 exemplifies its unwavering commitment to integrating state-of-the-art technologies, clinical trials, and community-driven approaches. By harnessing artificial intelligence, multiomics, and immunotherapy research, their scientists and clinicians are shaping the future landscape of cancer care, striving to transform hope into tangible cures for patients worldwide.

Subject of Research: Advances in cancer research and treatment integrating artificial intelligence, multiomics, and immunotherapy at City of Hope.

Web References:

• City of Hope
• AACR Abstracts Portal

Image Credits: City of Hope