Publications

Although species are central units for biological research, recent findings in genomics are raising awareness that what we call species can be ill-founded entities due to solely morphology-based, regional species descriptions. This particularly applies to groups characterized by intricate evolutionary processes such as hybridization, polyploidy, or asexuality. Here, challenges of current integrative taxonomy (genetics/genomics + morphology + ecology, etc.) become apparent: different favored species concepts, lack of universal characters/markers, missing appropriate analytical tools for intricate evolutionary processes, and highly subjective ranking and fusion of datasets. Now, integrative taxonomy combined with artificial intelligence under a unified species concept can enable automated feature learning and data integration, and thus reduce subjectivity in species delimitation. This approach will likely accelerate revising and unraveling eukaryotic biodiversity.

There is a lack of systematic research exploring cross-species variation in liver lobular geometry and zonation patterns of critical drug-metabolizing enzymes, a knowledge gap essential for translational studies. This study investigated the critical interplay between lobular geometry and key cytochrome P450 (CYP) zonation in four species: mouse, rat, pig, and human. We developed an automated pipeline based on whole slide images (WSI) of hematoxylin-eosin-stained liver sections and immunohistochemistry. This pipeline allows accurate quantification of both lobular geometry and zonation patterns of essential CYP proteins. Our analysis of CYP zonal expression shows that all CYP enzymes (besides CYP2D6 with panlobular expression) were observed in the pericentral region in all species, but with distinct differences. Comparison of normalized gradient intensity shows a high similarity between mice and humans, followed by rats. Specifically, CYP1A2 was expressed throughout the pericentral region in mice and humans, whereas it was restricted to a narrow pericentral rim in rats and showed a panlobular pattern in pigs. Similarly, CYP3A4 is present in the pericentral region, but its extent varies considerably in rats and appears panlobular in pigs. CYP2D6 zonal expression consistently shows a panlobular pattern in all species, although the intensity varies. CYP2E1 zonal expression covered the entire pericentral region with extension into the midzone in all four species, suggesting its potential for further cross-species analysis. Analysis of lobular geometry revealed an increase in lobular size with increasing species size, whereas lobular compactness was similar. Based on our results, zonated CYP expression in mice is most similar to humans. Therefore, mice appear to be the most appropriate species for drug metabolism studies unless larger species are required for other purposes, e.g., surgical reasons. CYP selection should be based on species, with CYP2E1 and CYP2D6 being the most preferable to compare four species. CYP1A2 could be considered as an additional CYP for rodent versus human comparisons, and CYP3A4 for mouse/human comparisons. In conclusion, our image analysis pipeline together with suggestions for species and CYP selection can serve to improve future cross-species and translational drug metabolism studies.

Modeling and simulating the growth of the branching of tree species remains a challenge. With existing approaches, we can reconstruct or rebuild the branching architectures of real tree species, but the simulation of the growth process remains unresolved. First, we present a tree growth model to generate branching architectures that resemble real tree species. Secondly, we use a quantitative morphometric approach to infer the shape similarity of the generated simulations and real tree species. Within a functional-structural plant model, we implement a set of biological parameters that affect the branching architecture of trees. By modifying the parameter values, we aim to generate basic shapes of spruce, pine, oak and poplar. Tree shapes are compared using geometric morphometrics of landmarks that capture crown and stem outline shapes. Five biological parameters, namely xylem flow, shedding rate, proprioception, gravitysense and lightsense, most influenced the generated tree branching patterns. Adjusting these five parameters resulted in the different tree shapes of spruce, pine, oak, and poplar. The largest effect was attributed to gravity, as phenotypic responses to this effect resulted in different growth directions of gymnosperm and angiosperm branching architectures. Since we were able to obtain branching architectures that resemble real tree species by adjusting only a few biological parameters, our model is extendable to other tree species. Furthermore, the model will also allow the simulation of structural tree-environment interactions. Our simplifying approach to shape comparison between tree species, landmark geometric morphometrics, showed that even the crown-trunk outlines capture species differences based on their contrasting branching architectures.

Gamification is widely used due to its positive influence on learning by adding emotions and steering behavior. In medical VR training applications, the use of gamification is rare, and when it is implemented, it often lacks thoughtful design decisions and empirical evaluation. Using a VR-based training for intraoperative ultrasound for liver surgery, we analyzed game elements regarding their suitability and examined two in more detail: difficulty levels and a kit, where the user has to assemble a virtual liver using US. In a broad audience study, levels achieved significantly better results regarding enjoyment. Qualitative feedback from medical students directly comparing the elements revealed that they prefer the kit as well as levels for training. Our studies indicate that levels and the more interactive kit improve the learning experience, which could also be taken as a basis for similar VR-based medical training applications.

Purpose: Age-related macular degeneration (AMD) is a complex eye disorder affecting millions worldwide. This article uses deep learning techniques to investigate the relationship between AMD, genetics and optical coherence tomography (OCT) scans. Methods: The cohort consisted of 332 patients, of which 235 were diagnosed with AMD and 97 were controls with no signs of AMD. The genome-wide association studies summary statistics utilized to establish the polygenic risk score (PRS) in relation to AMD were derived from the GERA European study. A PRS estimation based on OCT volumes for both eyes was performed using a proprietary convolutional neural network (CNN) model supported by machine learning models. The method's performance was assessed using numerical evaluation metrics, and the Grad-CAM technique was used to evaluate the results by visualizing the features learned by the model. Results: The best results were obtained with the CNN and the Extra Tree regressor (MAE = 0.55, MSE = 0.49, RMSE = 0.70, R2 = 0.34). Extending the feature vector with additional information on AMD diagnosis, age and smoking history improved the results slightly, with mainly AMD diagnosis used by the model (MAE = 0.54, MSE = 0.44, RMSE = 0.66, R2 = 0.42). Grad-CAM heatmap evaluation showed that the model decisions rely on retinal morphology factors relevant to AMD diagnosis. Conclusion: The developed method allows an efficient PRS estimation from OCT images. A new technique for analysing the association of OCT images with PRS of AMD, using a deep learning approach, may provide an opportunity to discover new associations between genotype-based AMD risk and retinal morphology.