Stem Cell Variation & Embryonic Development

The animation shows the mapping of this transcriptomic variation (as a color spectrum spanning the 1st principal component of bulk RNA-seq data from 317 hPSC lines derived from 101 human donors in Carcamo-Orive et al 2018) projected onto mouse embryonic development (the published UMAP of scRNA-seq data in Pijuan-Sala et al 2019). The low end of this primary dimension in hiPSC variation (purple & blue) spans pluripotency and the early neural lineage, while mesendodermal fates are on the high end (red & yellow). This is one of the final figures in our report that is currently under peer-review and available on biorXiv. The series of images at left are figure panels from our manuscript, and along with their captions here provide a brief guide through our discoveries. Click on an image and page through the slide show. You can explore all the novel and public datasets we use in this report at NeMO Analytics one gene at a time HERE, or via the transcriptome-wide expression patterns we dissected from bulk RNA-seq data in self-renewing and differentiating hPSC lines HERE. In addition, we have created a broader collection of public multi-omic data derived from mammalian embryos HERE. We encourage others with interest in this field to contribute their data to this integrated environment so that it can be further explored by those who generate the data as well as the wider research community. Please reach out through email or the CONNECT form below if you are interested in this and need guidance.

We have shown that variation in the lineage bias of pluripotent cells from different humans can be mapped onto the primary axes of in vivo mammalian development, and that this variation in cell potential is driven by stable transcriptomic signatures that are unique to all the cells of an individual human. These insights into the origins of human variation are instrumental in guiding how iPSC systems can be used more powerfully to develop disease models and novel therapeutics. In particular, in addition to distant end point assays examining mature cell types, we advocate the design of developmentally-inspired cell assays within pluripotency to interrogate large collections of hPSC lines from genetically diverse donors in order to understand human cellular variation and empower efforts in personalized medicine. We view these efforts as fueling an under appreciated goal: assays which reveal the inherent functional propensities of individal hiPSC lines by placing cells at particular moleclar decision points, i.e. assays which produce replicable results within individual hiPSC lines (and multiple lines from a single donor), but variable results across lines from distinct donors.

In this spirit, we are initiating a new virtal seminar series focused on stem cell models and variation across human iPCS lines: Developmental Diversity in a Dish (DDD). You can see Dr. Suel-Kee Kim present this work at the Jan. 9, 2024 session of the DDD seminar series HERE. Carlo Colantuoni also presented this work at the Feb. 09, 2023 Virtual Gastrulation Zoom Talk Series.

The collaborative team that generated this work began in Dr. Ron McKay’s Brain Development group at the Lieber Institute for Brain Development: Dr. Suel-kee Kim (now in Nenad Sestan’s Lab at Yale University), Dr. Seungmae Seo (now in Emily Mace’s lab at Columbia University), Dr. Genevieve Stein-O’Brien (now a PI in the Department of Neuroscience at Johns Hopkins School of Medicine), and Dr. Amritha Jaishankar (who lead the Maryland Stem Cell Research Fund, now at IQVIA). To elucidate these origins of human cellular variation, we combined stem cell expertise of Suel-Kee and others in Ron’s group with multi-omic data collection and integration via transfer learning techniques that Genevieve developed while a graduate student in my group. Collaborating with many members of this same team, we published a paper focused on the emergence of regional telencephalic organizer identities during in vitro neural differentiation, lead by Dr. Nicola Micali in Cell Reports.

Leveraging existing public data to understand novel emerging datasets in this way is central in our collaborative approach to discovery - you can check out our multi-omics data collections and exploration environment at NeMO Analytics.