Biology subjects

Stuck, M.

Publications and source records attributed to Stuck, M..

3 recordsLinked to original sources

bioRxiv2023

Burstprop for Learning in Spiking Neuromorphic Hardware

The need for energy-efficient solutions in Deep Neural Network (DNN) applications has led to a growing interest in Spiking Neural Networks (SNNs) implemented in neuromorphic hardware. The Burstprop algorithm enables online and local learning in hier-archical networks, and therefore can potentially be implemented in neuromorphic hardware. This work presents an adaptation of the algorithm for training hierarchical SNNs on MNIST. Our implementation requires an order of magnitude fewer neurons than the previous ones. While Burstprop outper-forms Spike-timing dependent plasticity (STDP), it falls short compared to training with backpropagation through time (BPTT). This work establishes a foundation for further improvements in the Burst-prop algorithm, developing such algorithms is essential for achieving energy-efficient machine learning in neuromorphic hardware.

neuroscience↗

bioRxiv2023

Direct translation of incoming retroviral RNA genomes

Viruses that carry a positive-sense, single-stranded (+ssRNA) RNA translate their genomes soon after entering the host cell to produce viral proteins, with the exception of retroviruses. A distinguishing feature of retroviruses is reverse transcription, where the +ssRNA genome serves as a template to synthesize a double-stranded DNA copy that subsequently integrates into the host genome. As retroviral RNAs are produced by the host cell transcriptional machinery and are largely indistinguishable from cellular mRNAs, we investigated the potential of incoming retroviral genomes to directly express proteins. Here we show through multiple, complementary methods that retroviral genomes are translated after entry. Our findings challenge the notion that retroviruses require reverse transcription to produce viral proteins. Synthesis of retroviral proteins in the absence of productive infection has significant implications for basic retrovirology, immune responses and gene therapy applications.

microbiology↗

bioRxiv2022

HIV-1 infection causes depletion of monocytic cells through a non-canonical cell death pathway

Programmed cell death is a regulatory mechanism to eliminate infected or damaged cells. Several programmed cell death pathways exist, including apoptosis, necroptosis and pyroptosis, which can be distinguished by the cellular molecules involved. Here we show that infection of monocytic cells with HIV-1 causes cell death, which is dose- and cell type dependent and occurs independently of nucleic acid sensing or interferon (IFN) signaling. Death is observed in the case of near full-length viruses that produce viral proteins upon infection, but not in case of a minimal lentiviral vector that does not express viral gene products, demonstrating the necessity of viral gene products or a near-full length RNA genome to trigger death. Inhibition of reverse transcription or integration rescues cells, indicating that a step after integration is responsible. Using mutant viruses, we further narrow down the step in the retroviral replication cycle that triggers death. Inhibition of diverse cell death pathways individually cannot rescue cells from death following infection, consistent with PANoptosis, a cellular death process that cannot be accounted for by any single programmed cell death pathway alone. Our results elucidate the viral and cellular determinants of cell death caused by HIV-1 infection and outline cellular responses that result in the depletion of specific cell populations.

microbiology↗