Planarians are among the most complex animals with the ability to regenerate complete organisms from small tissue pieces. This ability allows them to reproduce by splitting themselves into a head and a tail piece, making them a rare example of asexual reproduction via transverse fission in multi-cellular organisms. Due to the stochastic nature of long reproductive cycles, which range from days to months, few and primarily qualitative studies have been conducted to understand the reproductive behaviors of asexual planarians. We have executed the largest long-term study on planarian asexual reproduction to date, tracking more than 23,000 reproductive events of three common planarian species found in Europe, North America, and Asia, respectively: Schmidtea mediterranea, Dugesia tigrina, and Dugesia japonica. This unique data collection allowed us to perform a detailed statistical analysis of their reproductive strategies. Since the three species share a similar anatomy and mode of reproduction by transverse division, we were surprised to find that each species had acquired its own distinct strategy for optimizing its reproductive success. We statistically examined each strategy, associated trade-offs, and the potential regulatory mechanisms on the population level. Interestingly, models for cell cycle length regulation in unicellular organisms could be directly applied to describe reproductive cycle lengths of planarians, despite the difference in underlying biological mechanisms. Finally, we exam-Electronic supplementary material The online version of this article (ined the ecological implications of each strategy through intra-and inter-species competition experiments and found that D. japonica outcompeted the other two species due to its relatively equal distribution of resources on head and tail pieces, its cannibalistic behaviors and ability to thrive in crowded environments. These results show that this species would pose a serious threat to endogenous planarian populations if accidentally introduced in their habitats.
Planarians are an important model organism for regeneration and stem cell research. A complete understanding of stem cell and regeneration dynamics in these animals requires time-lapse imaging in vivo, which has been difficult to achieve due to a lack of tissue-specific markers and the strong negative phototaxis of planarians. We have developed the Planarian Immobilization Chip (PIC) for rapid, stable immobilization of planarians for in vivo imaging without injury or biochemical alteration. The chip is easy and inexpensive to fabricate, and worms can be mounted for and removed after imaging within minutes. We show that the PIC enables significantly higher-stability immobilization than can be achieved with standard techniques, allowing for imaging of planarians at sub-cellular resolution in vivo using brightfield and fluorescence microscopy. We validate the performance of the PIC by performing time-lapse imaging of planarian wound closure and sequential imaging over days of head regeneration. We further show that the device can be used to immobilize Hydra, another photophobic regenerative model organism. The simple fabrication, low cost, ease of use, and enhanced specimen stability of the PIC should enable its broad application to in vivo studies of stem cell and regeneration dynamics in planarians and Hydra.
Deep Neural Networks for image classification have been found to be vulnerable to adversarial samples, which consist of sub-perceptual noise added to a benign image that can easily fool trained neural networks, posing a significant risk to their commercial deployment.In this work, we analyze adversarial samples through the lens of their contributions to the principal components of each image, which is different than prior works in which authors performed PCA on the entire dataset. We investigate a number of state-of-the-art deep neural networks trained on ImageNet as well as several attacks for each of the networks. Our results demonstrate empirically that adversarial samples across several attacks have similar properties in their contributions to the principal components of neural network inputs. We propose a new metric for neural networks to measure their robustness to adversarial samples, termed the (k, p) point. We utilize this metric to achieve 93.36% accuracy in detecting adversarial samples independent of architecture and attack type for models trained on ImageNet.
Ultrafast terahertz spectroscopy can be used to probe charge and spin dynamics in semiconductors. We have studied THz emission from bulk InAs and GaAs and from GaAs/AlGaAs quantum wells as a function of magnetic field. Ultrashort pulses of THz radiation were produced at semiconductor surfaces by photoexcitation with a femtosecond TiSapphire laser, and we recorded the THz emission spectrum and the integrated THz power as a function of magnetic field and temperature. In bulk samples the emitted radiation is produced by coupled cyclotron-plasma oscillations: we model THz emission from n-GaAs as magneto-plasma oscillations in a 3-D electron gas. THz emission from a modulation-doped parabolic quantum well is described in terms of coupled intersubband-cyclotron motion. A model including both 3-D plasma oscillations and a 2-D electron gas in a surface accumulation layer is required to describe THz emission from InAs in a magnetic field.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.