Brooks, G. | Search bioRxiv

bioRxiv2024

Fundamental constraints on the evolution of vertebrate life histories

Vertebrate life histories evolve in response to selection imposed by abiotic and biotic environmental conditions while being limited by genetic, developmental, physiological, demographic, and phylogenetic processes that constrain adaptation. Despite the well-recognized shifts in selective pressures accompanying transitions among environments, identifying the conditions driving innovation and the consequences for life-history evolution remain an outstanding question. Here, we compare the traits of aquatic and terrestrial vertebrates to infer shifts in demographic and evolutionary constraints that explain differences in life-history optimization. Specifically, our results emphasize the reduced potential for life-history diversification on land, especially that of reproductive strategies. Moreover, our study reveals differences between the evolution of viviparity in the two realms. Transitions from egg laying to live birth represents a major shift across life-history space for aquatic organisms, whereas terrestrial egg-laying organisms evolve live birth without drastic changes in life-history strategy. Whilst trade-offs in the allocation of resources place fundamental constraints on the way life histories can vary, ecological setting influences the position of species within the viable phenotypic space available for adaptive evolution.

evolutionary biology↗

bioRxiv2022

Graph neural networks ameliorate potential impacts of imprecise large-scale autonomous immunofluorescence labeling of immune cells on whole slide images

The characteristics of tumor-infiltrating lymphocytes (TIL) are essential in cancer prognostication and treatment through the ability to indicate the tumors capacity to evade the immune system (e.g., as evidenced by nodal involvement). Machine learning technologies have demonstrated remarkable success for localizing TILs, though these methods require extensive curation of manual annotations or restaining procedures that can degrade tissue quality, resulting in imprecise annotation. In this study, we co-registered tissue slides stained for both hematoxylin and eosin (H&E) and immunofluorescence (IF) as means to rapidly perform large-scale annotation of nuclei. We integrated the following approaches to improve the prediction of TILs: 1) minimized tissue degradation on same-section tissue restaining, 2) developed a scoring algorithm to improve the selection of patches for machine learning modeling and 3) utilized a graph neural network deep learning approach to identify relevant contextual features for lymphocyte prediction. Our graph neural network approach accounts for surrounding contextual micro/macro-architecture tissue features to facilitate interpretation of registered IF. The graph neural network compares favorably (F1-score=0.9235, AUROC=0.9462) to two alternative modeling approaches. This study brings insight to the importance of contextual information leveraged from within and around neighboring cells in a nuclei classification workflow, as well as elucidate approaches which enable the rapid generation of large-scale annotations of lymphocytes for machine learning approaches for immune phenotyping. Such approaches can help further interrogate the spatial biology of colorectal cancer tumors and tumor metastasis.

pathology↗