Biology subjects

Horn, M.

Publications and source records attributed to Horn, M..

22 records · Page 2Linked to original sources

bioRxiv2020

Evolutionary recent dual obligatory symbiosis among adelgids indicates a transition between fungus and insect associated lifestyles

Adelgids (Insecta: Hemiptera: Adelgidae) form a small group of insects but harbor a surprisingly diverse set of bacteriocyte-associated endosymbionts, which suggest multiple replacement and acquisition of symbionts over evolutionary time. Specific pairs of symbionts have been associated with adelgid lineages specialized on different secondary host conifers. Using a metagenomic approach, we investigated the symbiosis of the Adelges laricis/tardus species complex containing betaproteobacterial ( Candidatus Vallotia tarda) and gammaproteobacterial ( Candidatus Profftia tarda) symbionts. Genomic characteristics and metabolic pathway reconstructions revealed that Vallotia and Profftia are evolutionary young endosymbionts, which complement each others role in essential amino acid production. Phylogenomic analyses and a high level of genomic synteny indicate an origin of the betaproteobacterial symbiont from endosymbionts of Rhizopus fungi. This evolutionary transition was accompanied with substantial loss of functions related to transcription regulation, secondary metabolite production, bacterial defense mechanisms, host infection and manipulation. The transition from fungus to insect endosymbionts extends our current framework about evolutionary trajectories of host-associated microbes.

microbiology↗

bioRxiv2020

Acquisition of a type 3 secretion signal in an housekeeping enzyme shaped glycogen metabolism in Chlamydia

The obligate intracellular bacteria Chlamydia trachomatis store glycogen in the lumen of the vacuoles in which they grow. Glycogen catabolism generates glucose-1-phosphate (Glc1P), while the bacteria are capable of taking up only glucose-6-phosphate (Glc6P). We tested whether the conversion of Glc1P into Glc6P could be catalyzed by a phosphoglucomutase (PGM) of host or bacterial origin. We found no evidence for the presence of the host enzyme in the vacuole. In C. trachomatis, two proteins are potential PGMs. By reconstituting the reaction, and by complementing PGM deficient fibroblasts, we demonstrated that only CT295 displayed robust PGM activity. Furthermore, we showed that glycogen accumulation by a subset of Chlamydia species correlated with the presence of a type three secretion (T3S) signal in their PGM. In conclusion, we established that the conversion of Glc1P into Glc6P was accomplished by a bacterial PGM, through the acquisition of a T3S signal in a "housekeeping" gene.

microbiology↗

bioRxiv2019

Natural Zeitgebers cannot compensate for the loss of a functional circadian clock in timing of a vital behaviour in Drosophila

The adaptive significance of adjusting behavioural activities to the right time of the day is intuitive. Laboratory studies have implicated an important role of circadian clocks in behavioural timing and rhythmicity. Yet, recent studies on clock-mutant animals questioned this importance under more naturalistic settings, as various clock mutants showed nearly normal diel activity rhythms under semi-natural Zeitgeber conditions. We here report evidence that proper timing of eclosion, a vital behaviour of the fruit fly Drosophila melanogaster, requires a functional molecular clock even under quasi-natural conditions. In contrast to wildtype flies, period01 mutants with a defective molecular clock eclose mostly arrhythmically in a temperate environment even in the presence of a full complement of abiotic Zeitgebers. Moreover, period01 mutants eclose during a much larger portion of the day, and peak eclosion time becomes more susceptible to variable day-to-day changes of light and temperature. Under the same conditions, flies with impaired peptidergic inter-clock signalling (pdf01 and han5304 mutants) stayed largely rhythmic with normal gate sizes. Our results suggest that the presence of natural Zeitgebers can mitigate a loss of peptide-mediated phasing between central clock neuron groups, but cannot substitute for the lack of a functional molecular clock under natural temperate conditions.

animal behavior and cognition↗

bioRxiv2018

Supervised learning on synthetic data for reverse engineering gene regulatory networks from experimental time-series

The reconstruction of gene regulatory networks from time resolved gene expression measurements is a key challenge in systems biology with applications in health and disease. While the most popular network inference methods are based on unsupervised learning approaches, supervised learning methods have proven their potential for superior reconstruction performance. However, obtaining the appropriate volume of informative training data constitutes a key limitation for the success of such methods.\n\nHere, we introduce a supervised learning approach to detect gene-gene regulation based on exclusively synthetic training data, termed surrogate learning, and show its performance for synthetic and experimental time-series. We systematically investigate different simulation configurations of biologically representative time-series of transcripts and augmentation of the data with a measurement model. We compare the resulting synthetic datasets to experimental data, and evaluate classifiers trained on them for detection of gene-gene regulation from experimental time-series. For classifiers, we consider hybrid convolutional recurrent neural networks, random forests and logistic regression, and evaluate the reconstruction performance of different simulation settings, data pre-processing and classifiers.\n\nWhen training and test time-courses are generated from the same distribution, we find that the largest tested neural network architecture achieves the best performance of 0.448 {+/-} 0.047 (mean {+/-} std) in maximally achievable F1 score over all datasets outperforming random forests by 32.4 % {+/-} 14 % (mean {+/-} std). Reconstruction performance is sensitive to discrepancies between synthetic training and test data, highlighting the importance of matching training and test data domains. For an experimental gene expression dataset from E.coli, we find that training data generated with measurement model, multi-gene perturbations, but without data standardization is best suited for training classifiers for network reconstruction from the experimental test data. We further demonstrate superiority to multiple unsupervised, state-of-the-art methods for networks comprising 20 genes of the experimental data from E.coli (average AUPR best supervised = 0.22 vs best unsupervised = 0.07).\n\nWe expect the proposed surrogate learning approach to be broadly applicable. It alleviates the requirement for large, difficult to attain volumes of experimental training data and instead relies on easily accessible synthetic data. Successful application for new experimental conditions and other data types is only limited by the automatable and scalable process of designing simulations which generate suitable synthetic data.

systems biology↗