Search bioRxiv⌕ Search
Home/Archive/bioRxiv

bioRxiv · 10.1101/2024.05.13.593993

OoCount: A machine-learning based approach to mouse ovarian follicle counting and classification

Folts, L.·Martinez, A. S.·Bunce, C.·Capel, B.·McKey, J.
View original source ↗Export metadata ↓

Abstract

The number and distribution of ovarian follicles in each growth stage provides a reliable readout of ovarian health and function. Leveraging techniques for three-dimensional (3D) imaging of ovaries in toto has the potential to uncover total, accurate ovarian follicle counts. However, because of the size and holistic nature of these images, counting oocytes is time consuming and difficult. The advent of deep-learning algorithms has allowed for the rapid development of ultra-fast, automated methods to analyze microscopy images. In recent years, these pipelines have become more user-friendly and accessible to non-specialists. We used these tools to create OoCount, a high-throughput, open-source method for automatic oocyte segmentation and classification from fluorescent 3D microscopy images of whole mouse ovaries using a deep-learning convolutional neural network (CNN) based approach. We developed a fast tissue-clearing and spinning disk confocal-based imaging protocol to obtain 3D images of whole mount perinatal and adult mouse ovaries. Fluorescently labeled oocytes from 3D images of ovaries were manually annotated in Napari to develop a machine learning training dataset. This dataset was used to retrain StarDist using a CNN within DL4MicEverywhere to automatically label all oocytes in the ovary. In a second phase, we utilize Accelerated Pixel and Object Classification, a Napari plugin, to classify labeled oocytes and sort them into growth stages. Here, we provide an end-to-end protocol for producing high-quality 3D images of the perinatal and adult mouse ovary, obtaining follicle counts and staging. We also demonstrate how to customize OoCount to fit images produced in any lab. Using OoCount, we can obtain accurate counts of oocytes in each growth stage in the perinatal and adult ovary, improving our ability to study ovarian function and fertility. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=136 SRC="FIGDIR/small/593993v1_ufig1.gif" ALT="Figure 1"> View larger version (29K): org.highwire.dtl.DTLVardef@8d7257org.highwire.dtl.DTLVardef@4dbe73org.highwire.dtl.DTLVardef@1283aaorg.highwire.dtl.DTLVardef@1822baf_HPS_FORMAT_FIGEXP M_FIG C_FIG Summary sentenceThis protocol introduces OoCount, a high-throughput, open-source method for automatic oocyte segmentation and classification from fluorescent 3D microscopy images of whole mouse ovaries using a machine learning-based approach.

Source connections

Explore related subjects

developmental biologyPreprint

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Folts, L., Martinez, A. S., Bunce, C., Capel, B., McKey, J.. 2024-05-14. OoCount: A machine-learning based approach to mouse ovarian follicle counting and classification. https://doi.org/10.1101/2024.05.13.593993

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Whole embryo barcoding reveals convergent origins of mesodermal cells during mammalian gastrulation

Classical fate-mapping and recent single-cell trajectory studies have suggested that mesoderm diversification during gastrulation proceeds through progressive fate bias and branching lineage segregation. Here, we combine inducible CRISPR-based lineage recording with single-cell RNA sequencing to reconstruct clonal relationships across mouse embryos barcoded at primitive streak onset and analyzed at early organogenesis. Mesodermal clones diverged after streak formation, but their developmental outputs remained extensively overlapping, revealing a clonal architecture that is not a simple branching lineage tree. Using sampling-aware statistical models, we quantify distinct clonal origins of paraxial mesoderm across the anterior-posterior axis, show that endothelial cells arise from multiple mesoderm origins but demonstrate transcriptional differences that correspond to anatomical position and clonal lineage, and identify clonal sharing between extra-embryonic mesoderm and embryonic derivatives, including cardiac lineages and a shared hindlimb-allantois progenitor. In the heart, progenitors corresponding to classical heart fields remain multipotent, contributing both myocardial and non-myocardial cell types. Together, our findings support a model in which mesodermal diversity is assembled through spatially distinct, fate-biased progenitor pools whose descendants converge on shared cell states. Our findings challenge simple branching models of mesoderm diversification and provide a clonal framework for understanding how tissue diversity emerges in the embryo.

developmental biology↗
bioRxiv2026

Transcription factor handover reshapes hub dynamics to coordinate temporal progression of Notch response

To decipher information in space and time, cells require mechanisms to sustain gene expression programmes once an initial signal decays. Localised condensation of transcription complexes into hubs is emerging as a key step in gene regulation and, by combining live imaging of transcription and of transcription complexes at target loci, we have investigated how these evolve to enable a prolonged transcription outcome in response to Notch signalling. Measuring the dynamics of hindsight (hnt) transcription reveals two phases. The first, a probabilistic response similar to E(spl)-C genes, is directly controlled by Notch and the second, a sustained response, is indirectly controlled through the BTB transcription factor Broad. The two phases correlate with a transition in transcription factor enrichment at hnt: complexes of Notch coactivator Mastermind and of Broad are successively enriched. Notably, the lifetime of the hubs formed by each are distinct and correlate with the change in transcription profiles. These reflect the binding dynamics of each complex, with Broad having significantly more stable association with chromatin, forming longer lasting hubs and yielding sustained transcription. Furthermore at intermediate stages we detect hubs containing both Broad and Mam implying an evolution in hub composition. Together our data support that the dynamic properties of transcription hubs are crucial for the probability and dynamics of transcription and point towards a ''hub handover'' in enabling temporal progression in gene expression responses.

developmental biology↗
bioRxiv2026

An interplay of multilayered regulatory mechanisms underlies the RNA solubility phase transition during vertebrate oocyte-to-embryo transition

Dynamic RNA-protein interactions influence RNA transport, storage, stability, and activity and are fundamental to a broad range of biological processes. During the oocyte-to-embryo transition (OET), fully differentiated gametes fuse and regain totipotency. Throughout the OET, the transcription remains silent. Spatiotemporal regulation of gene expression relies on the dynamic interactions between maternal transcripts and various protein complexes. Despite its fundamental importance, RNA-protein interaction dynamics at the global level during the OET remain poorly understood. Here, we leverage fractionation RNA-seq to profile RNA insolubility, a proxy for the interaction between RNAs and protein complexes, during zebrafish OET. We identify transcripts that transition between the soluble (relatively protein-free) and insoluble (heavily protein-associated) states and demonstrate that these phase transitions are driven by ribosome-dependent and ribosome-independent mechanisms. We demonstrate that the capacity of maternal RNAs to engage with protein complexes changes dynamically during the OET through remodeling of cis-regulatory features. We developed a machine-learning framework and uncovered multiple cis- and trans-regulatory features that may contribute to RNA insolubility. To validate these predictions, we examined poly(A) tail length and m6A dynamics. We found that poly(A) tail length is associated with both ribosome-dependent and ribosome-independent RNA phase transitions. In parallel, fractionation m6A-pulldown RNA-seq revealed highly dynamic changes in the m6A landscape during the OET. A correlation between m6A modifications and RNA insolubility is gradually established during the OET. Together, our findings uncover a multilayered regulatory mechanism governing RNA phase transitions during the OET, providing fundamental insights into post-transcriptional regulation of gene expression during development.

developmental biology↗