Search bioRxiv⌕ Search

Biology subjects

Kocher, M.

Publications and source records attributed to Kocher, M..

2 recordsLinked to original sources

Contribution of neurons that express fruitless and Clock transcription factors to behavioral rhythms and courtship

Animals need to integrate information across neuronal networks that direct reproductive behaviors and circadian rhythms. In Drosophila, the master regulatory transcription factors that direct courtship behaviors and circadian rhythms are co-expressed in a small set of neurons. In this study we investigate the role of these neurons in both males and females. We find sex-differences in the number of these fruitless and Clock -expressing neurons (fru {cap} Clk neurons) that is regulated by male-specific Fru. We assign the fru {cap} Clk neurons to the electron microscopy connectome that provides high resolution structural information. We also discover sex-differences in the number of fru-expressing neurons that are post-synaptic targets of Clk-expressing neurons, with more post-synaptic targets in males. When fru {cap} Clk neurons are activated or silenced, males have a shorter period length. Activation of fru {cap} Clk neurons also changes the rate a courtship behavior is performed. We find that activation and silencing fru {cap} Clk neurons impacts the molecular clock in the sLNv master pacemaker neurons, in a cell-nonautonomous manner. These results reveal how neurons that subserve the two processes, reproduction and circadian rhythms, can impact behavioral outcomes in a sex-specific manner.

neuroscience↗

Towards robust and generalizable super-resolution generative adversarial networks for magnetic resonance neuroimaging: a cross-population approach

Magnetic resonance imaging (MRI) is fundamental to neuroscience, where detailed structural brain scans improve clinical diagnoses and provide accurate neuroanatomical information. Apart from time-consuming scanning protocols, higher image resolution can be obtained with super resolution algorithms. We investigated the generalization abilities of Super Resolution Generative Adversarial Neural Networks (SRGANs) across different populations. T1-weighted scans from three large cohorts were used, spanning older subjects, newborns, and patients with brain tumor- or treatment-induced tissue changes. Upsampling quality was validated using synthetic and anatomical metrics. Models were first trained on each cohort, yielding high image quality and anatomical fidelity. When applied across cohorts, no artifacts were introduced by the SRGANs. SRGANs that were trained on a dataset combining all cohorts also did not induce any population-based artifacts. We showed that SRGANs provide a prime example of robust AI, where application on unseen populations did not introduce artifacts due to training data bias (e.g., insertion or removal of tumor-related signals and contrast inversion). This is an important step in the deployment of SRGANs in real-world settings.

neuroscience↗