SATINN: An automated neural network-based classification of testicular sections allows for high-throughput histopathology of mouse mutants

Abstract

Motivation: The mammalian testis is a complex organ with a hierarchical organization that changes smoothly and stereotypically over time in normal adults. While testis histology is already an invaluable tool for identifying and describing developmental differences in evolution and disease, methods for standardized, digital image analysis of testis are needed to expand the utility of this approach. Results: We developed SATINN (Software for Analysis of Testis Images with Neural Networks), a multi-level framework for automated analysis of multiplexed immunofluorescence images from mouse testis. This approach uses a convolutional neural network (CNN) to classify nuclei from seminiferous tubules into 7 distinct cell types with an accuracy of 94.2%. These cell classifications are then used in a second-level tubule CNN, which places seminiferous tubules into one of 7 distinct tubule stages with 90.4% accuracy. We further describe numerous cell- and tubule-level statistics that can be derived from wildtype testis. Finally, we demonstrate how the classifiers and derived statistics can be used to rapidly and precisely describe pathology by applying our methods to image data from two mutant mouse lines. Our results demonstrate the feasibility and potential of using computer-assisted analysis for testis histology, an area poised to evolve rapidly on the back of emerging, spatially-resolved genomic and proteomic technologies. Availability and implementation: Scripts to apply the methods described here are available from http://github.com/conradlab/SATINN.

Deletion of an evolutionarily conserved TAD boundary compromises spermatogenesis in mice

“…4D). Furthermore, automated image analysis of immunofluorescent testis cross-sections (two testis sections per Dmrtb1_B -/-(n=3) and Dmrtb1_B +/+ (n=5) animals; total of n= 3,464 tubules) using SATINN 23 (Fig. 4E) showed a 35% decrease in efficiency of spermatogenesis in homozygous mice (4.05 ± 0.048 to 2.63 ± 0.038, mean ± SE; p< 10 -5 ), matching the 28% decrease estimated by daily sperm production (Fig.…”

“…Primary antibodies were incubated overnight at 4°C, secondary antibodies (1:500, Invitrogen, A21202 and A21207) for 1 hour at room temperature and tissue was counterstained with 1 µg/mL Hoechst 33342. Whole-testis cross-sections were analyzed with the Software for Analysis of Testis Images with Neural Networks (SATINN 23 ) for a total of 3,464 tubules (n=1,164 B396_BOS -/and n=2,300…”