@inproceedings{8728, abstract = {Discrete-time Markov Chains (MCs) and Markov Decision Processes (MDPs) are two standard formalisms in system analysis. Their main associated quantitative objectives are hitting probabilities, discounted sum, and mean payoff. Although there are many techniques for computing these objectives in general MCs/MDPs, they have not been thoroughly studied in terms of parameterized algorithms, particularly when treewidth is used as the parameter. This is in sharp contrast to qualitative objectives for MCs, MDPs and graph games, for which treewidth-based algorithms yield significant complexity improvements. In this work, we show that treewidth can also be used to obtain faster algorithms for the quantitative problems. For an MC with n states and m transitions, we show that each of the classical quantitative objectives can be computed in O((n+m)⋅t2) time, given a tree decomposition of the MC with width t. Our results also imply a bound of O(κ⋅(n+m)⋅t2) for each objective on MDPs, where κ is the number of strategy-iteration refinements required for the given input and objective. Finally, we make an experimental evaluation of our new algorithms on low-treewidth MCs and MDPs obtained from the DaCapo benchmark suite. Our experiments show that on low-treewidth MCs and MDPs, our algorithms outperform existing well-established methods by one or more orders of magnitude.}, author = {Asadi, Ali and Chatterjee, Krishnendu and Goharshady, Amir Kafshdar and Mohammadi, Kiarash and Pavlogiannis, Andreas}, booktitle = {Automated Technology for Verification and Analysis}, isbn = {9783030591519}, issn = {1611-3349}, location = {Hanoi, Vietnam}, pages = {253--270}, publisher = {Springer Nature}, title = {{Faster algorithms for quantitative analysis of MCs and MDPs with small treewidth}}, doi = {10.1007/978-3-030-59152-6_14}, volume = {12302}, year = {2020}, } @inproceedings{8089, abstract = {We consider the classical problem of invariant generation for programs with polynomial assignments and focus on synthesizing invariants that are a conjunction of strict polynomial inequalities. We present a sound and semi-complete method based on positivstellensaetze, i.e. theorems in semi-algebraic geometry that characterize positive polynomials over a semi-algebraic set. On the theoretical side, the worst-case complexity of our approach is subexponential, whereas the worst-case complexity of the previous complete method (Kapur, ACA 2004) is doubly-exponential. Even when restricted to linear invariants, the best previous complexity for complete invariant generation is exponential (Colon et al, CAV 2003). On the practical side, we reduce the invariant generation problem to quadratic programming (QCLP), which is a classical optimization problem with many industrial solvers. We demonstrate the applicability of our approach by providing experimental results on several academic benchmarks. To the best of our knowledge, the only previous invariant generation method that provides completeness guarantees for invariants consisting of polynomial inequalities is (Kapur, ACA 2004), which relies on quantifier elimination and cannot even handle toy programs such as our running example.}, author = {Chatterjee, Krishnendu and Fu, Hongfei and Goharshady, Amir Kafshdar and Goharshady, Ehsan Kafshdar}, booktitle = {Proceedings of the 41st ACM SIGPLAN Conference on Programming Language Design and Implementation}, isbn = {9781450376136}, location = {London, United Kingdom}, pages = {672--687}, publisher = {Association for Computing Machinery}, title = {{Polynomial invariant generation for non-deterministic recursive programs}}, doi = {10.1145/3385412.3385969}, year = {2020}, } @inproceedings{7810, abstract = {Interprocedural data-flow analyses form an expressive and useful paradigm of numerous static analysis applications, such as live variables analysis, alias analysis and null pointers analysis. The most widely-used framework for interprocedural data-flow analysis is IFDS, which encompasses distributive data-flow functions over a finite domain. On-demand data-flow analyses restrict the focus of the analysis on specific program locations and data facts. This setting provides a natural split between (i) an offline (or preprocessing) phase, where the program is partially analyzed and analysis summaries are created, and (ii) an online (or query) phase, where analysis queries arrive on demand and the summaries are used to speed up answering queries. In this work, we consider on-demand IFDS analyses where the queries concern program locations of the same procedure (aka same-context queries). We exploit the fact that flow graphs of programs have low treewidth to develop faster algorithms that are space and time optimal for many common data-flow analyses, in both the preprocessing and the query phase. We also use treewidth to develop query solutions that are embarrassingly parallelizable, i.e. the total work for answering each query is split to a number of threads such that each thread performs only a constant amount of work. Finally, we implement a static analyzer based on our algorithms, and perform a series of on-demand analysis experiments on standard benchmarks. Our experimental results show a drastic speed-up of the queries after only a lightweight preprocessing phase, which significantly outperforms existing techniques.}, author = {Chatterjee, Krishnendu and Goharshady, Amir Kafshdar and Ibsen-Jensen, Rasmus and Pavlogiannis, Andreas}, booktitle = {European Symposium on Programming}, isbn = {9783030449131}, issn = {1611-3349}, location = {Dublin, Ireland}, pages = {112--140}, publisher = {Springer Nature}, title = {{Optimal and perfectly parallel algorithms for on-demand data-flow analysis}}, doi = {10.1007/978-3-030-44914-8_5}, volume = {12075}, year = {2020}, } @article{6918, abstract = {We consider the classic problem of Network Reliability. A network is given together with a source vertex, one or more target vertices, and probabilities assigned to each of the edges. Each edge of the network is operable with its associated probability and the problem is to determine the probability of having at least one source-to-target path that is entirely composed of operable edges. This problem is known to be NP-hard. We provide a novel scalable algorithm to solve the Network Reliability problem when the treewidth of the underlying network is small. We also show our algorithm’s applicability for real-world transit networks that have small treewidth, including the metro networks of major cities, such as London and Tokyo. Our algorithm leverages tree decompositions to shrink the original graph into much smaller graphs, for which reliability can be efficiently and exactly computed using a brute force method. To the best of our knowledge, this is the first exact algorithm for Network Reliability that can scale to handle real-world instances of the problem.}, author = {Goharshady, Amir Kafshdar and Mohammadi, Fatemeh}, issn = {0951-8320}, journal = {Reliability Engineering and System Safety}, publisher = {Elsevier}, title = {{An efficient algorithm for computing network reliability in small treewidth}}, doi = {10.1016/j.ress.2019.106665}, volume = {193}, year = {2020}, } @phdthesis{7525, abstract = {The medial habenula (MHb) is an evolutionary conserved epithalamic structure important for the modulation of emotional memory. It is involved in regulation of anxiety, compulsive behavior, addiction (nicotinic and opioid), sexual and feeding behavior. MHb receives inputs from septal regions and projects exclusively to the interpeduncular nucleus (IPN). Distinct sub-regions of the septum project to different subnuclei of MHb: the bed nucleus of anterior commissure projects to dorsal MHb and the triangular septum projects to ventral MHb. Furthermore, the dorsal and ventral MHb project to the lateral and rostral/central IPN, respectively. Importantly, these projections have unique features of prominent co-release of different neurotransmitters and requirement of a peculiar type of calcium channel for release. In general, synaptic neurotransmission requires an activity-dependent influx of Ca2+ into the presynaptic terminal through voltage-gated calcium channels. The calcium channel family most commonly involved in neurotransmitter release comprises three members, P/Q-, N- and R-type with Cav2.1, Cav2.2 and Cav2.3 subunits, respectively. In contrast to most CNS synapses that mainly express Cav2.1 and/or Cav2.2, MHb terminals in the IPN exclusively express Cav2.3. In other parts of the brain, such as the hippocampus, Cav2.3 is mostly located to postsynaptic elements. This unusual presynaptic location of Cav2.3 in the MHb-IPN pathway implies unique mechanisms of glutamate release in this pathway. One potential example of such uniqueness is the facilitation of release by GABAB receptor (GBR) activation. Presynaptic GBRs usually inhibit the release of neurotransmitters by inhibiting presynaptic calcium channels. MHb shows the highest expression levels of GBR in the brain. GBRs comprise two subunits, GABAB1 (GB1) and GABAB2 (GB2), and are associated with auxiliary subunits, called potassium channel tetramerization domain containing proteins (KCTD) 8, 12, 12b and 16. Among these four subunits, KCTD12b is exclusively expressed in ventral MHb, and KCTD8 shows the strongest expression in the whole MHb among other brain regions, indicating that KCTD8 and KCTD12b may be involved in the unique mechanisms of neurotransmitter release mediated by Cav2.3 and regulated by GBRs in this pathway. In the present study, we first verified that neurotransmission in both dorsal and ventral MHb-IPN pathways is mainly mediated by Cav2.3 using a selective blocker of R-type channels, SNX-482. We next found that baclofen, a GBR agonist, has facilitatory effects on release from ventral MHb terminal in rostral IPN, whereas it has inhibitory effects on release from dorsal MHb terminals in lateral IPN, indicating that KCTD12b expressed exclusively in ventral MHb may have a role in the facilitatory effects of GBR activation. In a heterologous expression system using HEK cells, we found that KCTD8 and KCTD12b but not KCTD12 directly bind with Cav2.3. Pre-embedding immunogold electron microscopy data show that Cav2.3 and KCTD12b are distributed most densely in presynaptic active zone in IPN with KCTD12b being present only in rostral/central but not lateral IPN, whereas GABAB, KCTD8 and KCTD12 are distributed most densely in perisynaptic sites with KCTD12 present more frequently in postsynaptic elements and only in rostral/central IPN. In freeze-fracture replica labelling, Cav2.3, KCTD8 and KCTD12b are co-localized with each other in the same active zone indicating that they may form complexes regulating vesicle release in rostral IPN. On electrophysiological studies of wild type (WT) mice, we found that paired-pulse ratio in rostral IPN of KCTD12b knock-out (KO) mice is lower than those of WT and KCTD8 KO mice. Consistent with this finding, in mean variance analysis, release probability in rostral IPN of KCTD12b KO mice is higher than that of WT and KCTD8 KO mice. Although paired-pulse ratios are not different between WT and KCTD8 KO mice, the mean variance analysis revealed significantly lower release probability in rostral IPN of KCTD8 KO than WT mice. These results demonstrate bidirectional regulation of Cav2.3-mediated release by KCTD8 and KCTD12b without GBR activation in rostral IPN. Finally, we examined the baclofen effects in rostral IPN of KCTD8 and KCTD12b KO mice, and found the facilitation of release remained in both KO mice, indicating that the peculiar effects of the GBR activation in this pathway do not depend on the selective expression of these KCTD subunits in ventral MHb. However, we found that presynaptic potentiation of evoked EPSC amplitude by baclofen falls to baseline after washout faster in KCTD12b KO mice than WT, KCTD8 KO and KCTD8/12b double KO mice. This result indicates that KCTD12b is involved in sustained potentiation of vesicle release by GBR activation, whereas KCTD8 is involved in its termination in the absence of KCTD12b. Consistent with these functional findings, replica labelling revealed an increase in density of KCTD8, but not Cav2.3 or GBR at active zone in rostral IPN of KCTD12b KO mice compared with that of WT mice, suggesting that increased association of KCTD8 with Cav2.3 facilitates the release probability and termination of the GBR effect in the absence of KCTD12b. In summary, our study provided new insights into the physiological roles of presynaptic Cav2.3, GBRs and their auxiliary subunits KCTDs at an evolutionary conserved neuronal circuit. Future studies will be required to identify the exact molecular mechanism underlying the GBR-mediated presynaptic potentiation on ventral MHb terminals. It remains to be determined whether the prominent presence of presynaptic KCTDs at active zone could exert similar neuromodulatory functions in different pathways of the brain. }, author = {Bhandari, Pradeep}, issn = {2663-337X}, keywords = {Cav2.3, medial habenula (MHb), interpeduncular nucleus (IPN)}, pages = {79}, publisher = {Institute of Science and Technology Austria}, title = {{Localization and functional role of Cav2.3 in the medial habenula to interpeduncular nucleus pathway}}, doi = {10.15479/AT:ISTA:7525}, year = {2020}, } @article{7426, abstract = {This paper presents a novel abstraction technique for analyzing Lyapunov and asymptotic stability of polyhedral switched systems. A polyhedral switched system is a hybrid system in which the continuous dynamics is specified by polyhedral differential inclusions, the invariants and guards are specified by polyhedral sets and the switching between the modes do not involve reset of variables. A finite state weighted graph abstracting the polyhedral switched system is constructed from a finite partition of the state–space, such that the satisfaction of certain graph conditions, such as the absence of cycles with product of weights on the edges greater than (or equal) to 1, implies the stability of the system. However, the graph is in general conservative and hence, the violation of the graph conditions does not imply instability. If the analysis fails to establish stability due to the conservativeness in the approximation, a counterexample (cycle with product of edge weights greater than or equal to 1) indicating a potential reason for the failure is returned. Further, a more precise approximation of the switched system can be constructed by considering a finer partition of the state–space in the construction of the finite weighted graph. We present experimental results on analyzing stability of switched systems using the above method.}, author = {Garcia Soto, Miriam and Prabhakar, Pavithra}, issn = {1751-570X}, journal = {Nonlinear Analysis: Hybrid Systems}, number = {5}, publisher = {Elsevier}, title = {{Abstraction based verification of stability of polyhedral switched systems}}, doi = {10.1016/j.nahs.2020.100856}, volume = {36}, year = {2020}, } @article{7161, abstract = {In this paper, we introduce an inertial projection-type method with different updating strategies for solving quasi-variational inequalities with strongly monotone and Lipschitz continuous operators in real Hilbert spaces. Under standard assumptions, we establish different strong convergence results for the proposed algorithm. Primary numerical experiments demonstrate the potential applicability of our scheme compared with some related methods in the literature.}, author = {Shehu, Yekini and Gibali, Aviv and Sagratella, Simone}, issn = {1573-2878}, journal = {Journal of Optimization Theory and Applications}, pages = {877–894}, publisher = {Springer Nature}, title = {{Inertial projection-type methods for solving quasi-variational inequalities in real Hilbert spaces}}, doi = {10.1007/s10957-019-01616-6}, volume = {184}, year = {2020}, } @article{8002, abstract = {Wound healing in plant tissues, consisting of rigid cell wall-encapsulated cells, represents a considerable challenge and occurs through largely unknown mechanisms distinct from those in animals. Owing to their inability to migrate, plant cells rely on targeted cell division and expansion to regenerate wounds. Strict coordination of these wound-induced responses is essential to ensure efficient, spatially restricted wound healing. Single-cell tracking by live imaging allowed us to gain mechanistic insight into the wound perception and coordination of wound responses after laser-based wounding in Arabidopsis root. We revealed a crucial contribution of the collapse of damaged cells in wound perception and detected an auxin increase specific to cells immediately adjacent to the wound. This localized auxin increase balances wound-induced cell expansion and restorative division rates in a dose-dependent manner, leading to tumorous overproliferation when the canonical TIR1 auxin signaling is disrupted. Auxin and wound-induced turgor pressure changes together also spatially define the activation of key components of regeneration, such as the transcription regulator ERF115. Our observations suggest that the wound signaling involves the sensing of collapse of damaged cells and a local auxin signaling activation to coordinate the downstream transcriptional responses in the immediate wound vicinity.}, author = {Hörmayer, Lukas and Montesinos López, Juan C and Marhavá, Petra and Benková, Eva and Yoshida, Saiko and Friml, Jiří}, issn = {1091-6490}, journal = {Proceedings of the National Academy of Sciences of the United States of America}, number = {26}, publisher = {National Academy of Sciences}, title = {{Wounding-induced changes in cellular pressure and localized auxin signalling spatially coordinate restorative divisions in roots}}, doi = {10.1073/pnas.2003346117}, volume = {117}, year = {2020}, } @phdthesis{7258, abstract = {Many flows encountered in nature and applications are characterized by a chaotic motion known as turbulence. Turbulent flows generate intense friction with pipe walls and are responsible for considerable amounts of energy losses at world scale. The nature of turbulent friction and techniques aimed at reducing it have been subject of extensive research over the last century, but no definite answer has been found yet. In this thesis we show that in pipes at moderate turbulent Reynolds numbers friction is better described by the power law first introduced by Blasius and not by the Prandtl–von Kármán formula. At higher Reynolds numbers, large scale motions gradually become more important in the flow and can be related to the change in scaling of friction. Next, we present a series of new techniques that can relaminarize turbulence by suppressing a key mechanism that regenerates it at walls, the lift–up effect. In addition, we investigate the process of turbulence decay in several experiments and discuss the drag reduction potential. Finally, we examine the behavior of friction under pulsating conditions inspired by the human heart cycle and we show that under such circumstances turbulent friction can be reduced to produce energy savings.}, author = {Scarselli, Davide}, issn = {2663-337X}, pages = {174}, publisher = {Institute of Science and Technology Austria}, title = {{New approaches to reduce friction in turbulent pipe flow}}, doi = {10.15479/AT:ISTA:7258}, year = {2020}, } @phdthesis{8657, abstract = {Synthesis of proteins – translation – is a fundamental process of life. Quantitative studies anchor translation into the context of bacterial physiology and reveal several mathematical relationships, called “growth laws,” which capture physiological feedbacks between protein synthesis and cell growth. Growth laws describe the dependency of the ribosome abundance as a function of growth rate, which can change depending on the growth conditions. Perturbations of translation reveal that bacteria employ a compensatory strategy in which the reduced translation capability results in increased expression of the translation machinery. Perturbations of translation are achieved in various ways; clinically interesting is the application of translation-targeting antibiotics – translation inhibitors. The antibiotic effects on bacterial physiology are often poorly understood. Bacterial responses to two or more simultaneously applied antibiotics are even more puzzling. The combined antibiotic effect determines the type of drug interaction, which ranges from synergy (the effect is stronger than expected) to antagonism (the effect is weaker) and suppression (one of the drugs loses its potency). In the first part of this work, we systematically measure the pairwise interaction network for translation inhibitors that interfere with different steps in translation. We find that the interactions are surprisingly diverse and tend to be more antagonistic. To explore the underlying mechanisms, we begin with a minimal biophysical model of combined antibiotic action. We base this model on the kinetics of antibiotic uptake and binding together with the physiological response described by the growth laws. The biophysical model explains some drug interactions, but not all; it specifically fails to predict suppression. In the second part of this work, we hypothesize that elusive suppressive drug interactions result from the interplay between ribosomes halted in different stages of translation. To elucidate this putative mechanism of drug interactions between translation inhibitors, we generate translation bottlenecks genetically using in- ducible control of translation factors that regulate well-defined translation cycle steps. These perturbations accurately mimic antibiotic action and drug interactions, supporting that the interplay of different translation bottlenecks partially causes these interactions. We extend this approach by varying two translation bottlenecks simultaneously. This approach reveals the suppression of translocation inhibition by inhibited translation. We rationalize this effect by modeling dense traffic of ribosomes that move on transcripts in a translation factor-mediated manner. This model predicts a dissolution of traffic jams caused by inhibited translocation when the density of ribosome traffic is reduced by lowered initiation. We base this model on the growth laws and quantitative relationships between different translation and growth parameters. In the final part of this work, we describe a set of tools aimed at quantification of physiological and translation parameters. We further develop a simple model that directly connects the abundance of a translation factor with the growth rate, which allows us to extract physiological parameters describing initiation. We demonstrate the development of tools for measuring translation rate. This thesis showcases how a combination of high-throughput growth rate mea- surements, genetics, and modeling can reveal mechanisms of drug interactions. Furthermore, by a gradual transition from combinations of antibiotics to precise genetic interventions, we demonstrated the equivalency between genetic and chemi- cal perturbations of translation. These findings tile the path for quantitative studies of antibiotic combinations and illustrate future approaches towards the quantitative description of translation.}, author = {Kavcic, Bor}, isbn = {978-3-99078-011-4}, issn = {2663-337X}, pages = {271}, publisher = {Institute of Science and Technology Austria}, title = {{Perturbations of protein synthesis: from antibiotics to genetics and physiology}}, doi = {10.15479/AT:ISTA:8657}, year = {2020}, } @phdthesis{8620, abstract = {The development of the human brain occurs through a tightly regulated series of dynamic and adaptive processes during prenatal and postnatal life. A disruption of this strictly orchestrated series of events can lead to a number of neurodevelopmental conditions, including Autism Spectrum Disorders (ASDs). ASDs are a very common, etiologically and phenotypically heterogeneous group of disorders sharing the core symptoms of social interaction and communication deficits and restrictive and repetitive interests and behaviors. They are estimated to affect one in 59 individuals in the U.S. and, over the last three decades, mutations in more than a hundred genetic loci have been convincingly linked to ASD pathogenesis. Yet, for the vast majority of these ASD-risk genes their role during brain development and precise molecular function still remain elusive. De novo loss of function mutations in the ubiquitin ligase-encoding gene Cullin 3 (CUL3) lead to ASD. In the study described here, we used Cul3 mouse models to evaluate the consequences of Cul3 mutations in vivo. Our results show that Cul3 heterozygous knockout mice exhibit deficits in motor coordination as well as ASD-relevant social and cognitive impairments. Cul3+/-, Cul3+/fl Emx1-Cre and Cul3fl/fl Emx1-Cre mutant brains display cortical lamination abnormalities due to defective migration of post-mitotic excitatory neurons, as well as reduced numbers of excitatory and inhibitory neurons. In line with the observed abnormal cortical organization, Cul3 heterozygous deletion is associated with decreased spontaneous excitatory and inhibitory activity in the cortex. At the molecular level we show that Cul3 regulates cytoskeletal and adhesion protein abundance in the mouse embryonic cortex. Abnormal regulation of cytoskeletal proteins in Cul3 mutant neural cells results in atypical organization of the actin mesh at the cell leading edge. Of note, heterozygous deletion of Cul3 in adult mice does not induce the majority of the behavioral defects observed in constitutive Cul3 haploinsufficient animals, pointing to a critical time-window for Cul3 deficiency. In conclusion, our data indicate that Cul3 plays a critical role in the regulation of cytoskeletal proteins and neuronal migration. ASD-associated defects and behavioral abnormalities are primarily due to dosage sensitive Cul3 functions at early brain developmental stages.}, author = {Morandell, Jasmin}, issn = {2663-337X}, pages = {138}, publisher = {Institute of Science and Technology Austria}, title = {{Illuminating the role of Cul3 in autism spectrum disorder pathogenesis}}, doi = {10.15479/AT:ISTA:8620}, year = {2020}, } @article{8569, abstract = {Concerted radial migration of newly born cortical projection neurons, from their birthplace to their final target lamina, is a key step in the assembly of the cerebral cortex. The cellular and molecular mechanisms regulating the specific sequential steps of radial neuronal migration in vivo are however still unclear, let alone the effects and interactions with the extracellular environment. In any in vivo context, cells will always be exposed to a complex extracellular environment consisting of (1) secreted factors acting as potential signaling cues, (2) the extracellular matrix, and (3) other cells providing cell–cell interaction through receptors and/or direct physical stimuli. Most studies so far have described and focused mainly on intrinsic cell-autonomous gene functions in neuronal migration but there is accumulating evidence that non-cell-autonomous-, local-, systemic-, and/or whole tissue-wide effects substantially contribute to the regulation of radial neuronal migration. These non-cell-autonomous effects may differentially affect cortical neuron migration in distinct cellular environments. However, the cellular and molecular natures of such non-cell-autonomous mechanisms are mostly unknown. Furthermore, physical forces due to collective migration and/or community effects (i.e., interactions with surrounding cells) may play important roles in neocortical projection neuron migration. In this concise review, we first outline distinct models of non-cell-autonomous interactions of cortical projection neurons along their radial migration trajectory during development. We then summarize experimental assays and platforms that can be utilized to visualize and potentially probe non-cell-autonomous mechanisms. Lastly, we define key questions to address in the future.}, author = {Hansen, Andi H and Hippenmeyer, Simon}, issn = {2296-634X}, journal = {Frontiers in Cell and Developmental Biology}, number = {9}, publisher = {Frontiers}, title = {{Non-cell-autonomous mechanisms in radial projection neuron migration in the developing cerebral cortex}}, doi = {10.3389/fcell.2020.574382}, volume = {8}, year = {2020}, } @unpublished{7800, abstract = {De novo loss of function mutations in the ubiquitin ligase-encoding gene Cullin3 (CUL3) lead to autism spectrum disorder (ASD). Here, we used Cul3 mouse models to evaluate the consequences of Cul3 mutations in vivo. Our results show that Cul3 haploinsufficient mice exhibit deficits in motor coordination as well as ASD-relevant social and cognitive impairments. Cul3 mutant brain displays cortical lamination abnormalities due to defective neuronal migration and reduced numbers of excitatory and inhibitory neurons. In line with the observed abnormal columnar organization, Cul3 haploinsufficiency is associated with decreased spontaneous excitatory and inhibitory activity in the cortex. At the molecular level, employing a quantitative proteomic approach, we show that Cul3 regulates cytoskeletal and adhesion protein abundance in mouse embryos. Abnormal regulation of cytoskeletal proteins in Cul3 mutant neuronal cells results in atypical organization of the actin mesh at the cell leading edge, likely causing the observed migration deficits. In contrast to these important functions early in development, Cul3 deficiency appears less relevant at adult stages. In fact, induction of Cul3 haploinsufficiency in adult mice does not result in the behavioral defects observed in constitutive Cul3 haploinsufficient animals. Taken together, our data indicate that Cul3 has a critical role in the regulation of cytoskeletal proteins and neuronal migration and that ASD-associated defects and behavioral abnormalities are primarily due to Cul3 functions at early developmental stages.}, author = {Morandell, Jasmin and Schwarz, Lena A and Basilico, Bernadette and Tasciyan, Saren and Nicolas, Armel and Sommer, Christoph M and Kreuzinger, Caroline and Knaus, Lisa and Dobler, Zoe and Cacci, Emanuele and Danzl, Johann G and Novarino, Gaia}, booktitle = {bioRxiv}, title = {{Cul3 regulates cytoskeleton protein homeostasis and cell migration during a critical window of brain development}}, doi = {10.1101/2020.01.10.902064 }, year = {2020}, } @article{8250, abstract = {Antibiotics that interfere with translation, when combined, interact in diverse and difficult-to-predict ways. Here, we explain these interactions by “translation bottlenecks”: points in the translation cycle where antibiotics block ribosomal progression. To elucidate the underlying mechanisms of drug interactions between translation inhibitors, we generate translation bottlenecks genetically using inducible control of translation factors that regulate well-defined translation cycle steps. These perturbations accurately mimic antibiotic action and drug interactions, supporting that the interplay of different translation bottlenecks causes these interactions. We further show that growth laws, combined with drug uptake and binding kinetics, enable the direct prediction of a large fraction of observed interactions, yet fail to predict suppression. However, varying two translation bottlenecks simultaneously supports that dense traffic of ribosomes and competition for translation factors account for the previously unexplained suppression. These results highlight the importance of “continuous epistasis” in bacterial physiology.}, author = {Kavcic, Bor and Tkačik, Gašper and Bollenbach, Tobias}, issn = {2041-1723}, journal = {Nature Communications}, publisher = {Springer Nature}, title = {{Mechanisms of drug interactions between translation-inhibiting antibiotics}}, doi = {10.1038/s41467-020-17734-z}, volume = {11}, year = {2020}, } @article{8131, abstract = {The possibility to generate construct valid animal models enabled the development and testing of therapeutic strategies targeting the core features of autism spectrum disorders (ASDs). At the same time, these studies highlighted the necessity of identifying sensitive developmental time windows for successful therapeutic interventions. Animal and human studies also uncovered the possibility to stratify the variety of ASDs in molecularly distinct subgroups, potentially facilitating effective treatment design. Here, we focus on the molecular pathways emerging as commonly affected by mutations in diverse ASD-risk genes, on their role during critical windows of brain development and the potential treatments targeting these biological processes.}, author = {Basilico, Bernadette and Morandell, Jasmin and Novarino, Gaia}, issn = {1879-0380}, journal = {Current Opinion in Genetics and Development}, number = {12}, pages = {126--137}, publisher = {Elsevier}, title = {{Molecular mechanisms for targeted ASD treatments}}, doi = {10.1016/j.gde.2020.06.004}, volume = {65}, year = {2020}, } @unpublished{7673, abstract = {Combining drugs can improve the efficacy of treatments. However, predicting the effect of drug combinations is still challenging. The combined potency of drugs determines the drug interaction, which is classified as synergistic, additive, antagonistic, or suppressive. While probabilistic, non-mechanistic models exist, there is currently no biophysical model that can predict antibiotic interactions. Here, we present a physiologically relevant model of the combined action of antibiotics that inhibit protein synthesis by targeting the ribosome. This model captures the kinetics of antibiotic binding and transport, and uses bacterial growth laws to predict growth in the presence of antibiotic combinations. We find that this biophysical model can produce all drug interaction types except suppression. We show analytically that antibiotics which cannot bind to the ribosome simultaneously generally act as substitutes for one another, leading to additive drug interactions. Previously proposed null expectations for higher-order drug interactions follow as a limiting case of our model. We further extend the model to include the effects of direct physical or allosteric interactions between individual drugs on the ribosome. Notably, such direct interactions profoundly change the combined drug effect, depending on the kinetic parameters of the drugs used. The model makes additional predictions for the effects of resistance genes on drug interactions and for interactions between ribosome-targeting antibiotics and antibiotics with other targets. These findings enhance our understanding of the interplay between drug action and cell physiology and are a key step toward a general framework for predicting drug interactions.}, author = {Kavcic, Bor and Tkačik, Gašper and Bollenbach, Tobias}, booktitle = {bioRxiv}, title = {{A minimal biophysical model of combined antibiotic action}}, doi = {10.1101/2020.04.18.047886}, year = {2020}, } @unpublished{9750, abstract = {Tension of the actomyosin cell cortex plays a key role in determining cell-cell contact growth and size. The level of cortical tension outside of the cell-cell contact, when pulling at the contact edge, scales with the total size to which a cell-cell contact can grow1,2. Here we show in zebrafish primary germ layer progenitor cells that this monotonic relationship only applies to a narrow range of cortical tension increase, and that above a critical threshold, contact size inversely scales with cortical tension. This switch from cortical tension increasing to decreasing progenitor cell-cell contact size is caused by cortical tension promoting E-cadherin anchoring to the actomyosin cytoskeleton, thereby increasing clustering and stability of E-cadherin at the contact. Once tension-mediated E-cadherin stabilization at the contact exceeds a critical threshold level, the rate by which the contact expands in response to pulling forces from the cortex sharply drops, leading to smaller contacts at physiologically relevant timescales of contact formation. Thus, the activity of cortical tension in expanding cell-cell contact size is limited by tension stabilizing E-cadherin-actin complexes at the contact.}, author = {Slovakova, Jana and Sikora, Mateusz K and Caballero Mancebo, Silvia and Krens, Gabriel and Kaufmann, Walter and Huljev, Karla and Heisenberg, Carl-Philipp J}, booktitle = {bioRxiv}, pages = {41}, title = {{Tension-dependent stabilization of E-cadherin limits cell-cell contact expansion}}, doi = {10.1101/2020.11.20.391284}, year = {2020}, } @article{8532, abstract = {The molecular anatomy of synapses defines their characteristics in transmission and plasticity. Precise measurements of the number and distribution of synaptic proteins are important for our understanding of synapse heterogeneity within and between brain regions. Freeze–fracture replica immunogold electron microscopy enables us to analyze them quantitatively on a two-dimensional membrane surface. Here, we introduce Darea software, which utilizes deep learning for analysis of replica images and demonstrate its usefulness for quick measurements of the pre- and postsynaptic areas, density and distribution of gold particles at synapses in a reproducible manner. We used Darea for comparing glutamate receptor and calcium channel distributions between hippocampal CA3-CA1 spine synapses on apical and basal dendrites, which differ in signaling pathways involved in synaptic plasticity. We found that apical synapses express a higher density of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors and a stronger increase of AMPA receptors with synaptic size, while basal synapses show a larger increase in N-methyl-D-aspartate (NMDA) receptors with size. Interestingly, AMPA and NMDA receptors are segregated within postsynaptic sites and negatively correlated in density among both apical and basal synapses. In the presynaptic sites, Cav2.1 voltage-gated calcium channels show similar densities in apical and basal synapses with distributions consistent with an exclusion zone model of calcium channel-release site topography.}, author = {Kleindienst, David and Montanaro-Punzengruber, Jacqueline-Claire and Bhandari, Pradeep and Case, Matthew J and Fukazawa, Yugo and Shigemoto, Ryuichi}, issn = {1422-0067}, journal = {International Journal of Molecular Sciences}, number = {18}, publisher = {MDPI}, title = {{Deep learning-assisted high-throughput analysis of freeze-fracture replica images applied to glutamate receptors and calcium channels at hippocampal synapses}}, doi = {10.3390/ijms21186737}, volume = {21}, year = {2020}, } @inproceedings{9633, abstract = {The search for biologically faithful synaptic plasticity rules has resulted in a large body of models. They are usually inspired by – and fitted to – experimental data, but they rarely produce neural dynamics that serve complex functions. These failures suggest that current plasticity models are still under-constrained by existing data. Here, we present an alternative approach that uses meta-learning to discover plausible synaptic plasticity rules. Instead of experimental data, the rules are constrained by the functions they implement and the structure they are meant to produce. Briefly, we parameterize synaptic plasticity rules by a Volterra expansion and then use supervised learning methods (gradient descent or evolutionary strategies) to minimize a problem-dependent loss function that quantifies how effectively a candidate plasticity rule transforms an initially random network into one with the desired function. We first validate our approach by re-discovering previously described plasticity rules, starting at the single-neuron level and “Oja’s rule”, a simple Hebbian plasticity rule that captures the direction of most variability of inputs to a neuron (i.e., the first principal component). We expand the problem to the network level and ask the framework to find Oja’s rule together with an anti-Hebbian rule such that an initially random two-layer firing-rate network will recover several principal components of the input space after learning. Next, we move to networks of integrate-and-fire neurons with plastic inhibitory afferents. We train for rules that achieve a target firing rate by countering tuned excitation. Our algorithm discovers a specific subset of the manifold of rules that can solve this task. Our work is a proof of principle of an automated and unbiased approach to unveil synaptic plasticity rules that obey biological constraints and can solve complex functions.}, author = {Confavreux, Basile J and Zenke, Friedemann and Agnes, Everton J. and Lillicrap, Timothy and Vogels, Tim P}, booktitle = {34th Conference on Neural Information Processing Systems}, issn = {1049-5258}, location = {Vancouver, Canada}, pages = {16398--16408}, publisher = {Neural Information Processing Systems Foundation}, title = {{A meta-learning approach to (re)discover plasticity rules that carve a desired function into a neural network}}, volume = {33}, year = {2020}, } @phdthesis{7680, abstract = {Proteins and their complex dynamic interactions regulate cellular mechanisms from sensing and transducing extracellular signals, to mediating genetic responses, and sustaining or changing cell morphology. To manipulate these protein-protein interactions (PPIs) that govern the behavior and fate of cells, synthetically constructed, genetically encoded tools provide the means to precisely target proteins of interest (POIs), and control their subcellular localization and activity in vitro and in vivo. Ideal synthetic tools react to an orthogonal cue, i.e. a trigger that does not activate any other endogenous process, thereby allowing manipulation of the POI alone. In optogenetics, naturally occurring photosensory domain from plants, algae and bacteria are re-purposed and genetically fused to POIs. Illumination with light of a specific wavelength triggers a conformational change that can mediate PPIs, such as dimerization or oligomerization. By using light as a trigger, these tools can be activated with high spatial and temporal precision, on subcellular and millisecond scales. Chemogenetic tools consist of protein domains that recognize and bind small molecules. By genetic fusion to POIs, these domains can mediate PPIs upon addition of their specific ligands, which are often synthetically designed to provide highly specific interactions and exhibit good bioavailability. Most optogenetic tools to mediate PPIs are based on well-studied photoreceptors responding to red, blue or near-UV light, leaving a striking gap in the green band of the visible light spectrum. Among both optogenetic and chemogenetic tools, there is an abundance of methods to induce PPIs, but tools to disrupt them require UV illumination, rely on covalent linkage and subsequent enzymatic cleavage or initially result in protein clustering of unknown stoichiometry. This work describes how the recently structurally and photochemically characterized green-light responsive cobalamin-binding domains (CBDs) from bacterial transcription factors were re-purposed to function as a green-light responsive optogenetic tool. In contrast to previously engineered optogenetic tools, CBDs do not induce PPI, but rather confer a PPI already upon expression, which can be rapidly disrupted by illumination. This was employed to mimic inhibition of constitutive activity of a growth factor receptor, and successfully implement for cell signalling in mammalian cells and in vivo to rescue development in zebrafish. This work further describes the development and application of a chemically induced de-dimerizer (CDD) based on a recently identified and structurally described bacterial oxyreductase. CDD forms a dimer upon expression in absence of its cofactor, the flavin derivative F420. Safety and of domain expression and ligand exposure are demonstrated in vitro and in vivo in zebrafish. The system is further applied to inhibit cell signalling output from a chimeric receptor upon F420 treatment. CBDs and CDD expand the repertoire of synthetic tools by providing novel mechanisms of mediating PPIs, and by recognizing previously not utilized cues. In the future, they can readily be combined with existing synthetic tools to functionally manipulate PPIs in vitro and in vivo.}, author = {Kainrath, Stephanie}, issn = {2663-337X}, pages = {98}, publisher = {Institute of Science and Technology Austria}, title = {{Synthetic tools for optogenetic and chemogenetic inhibition of cellular signals}}, doi = {10.15479/AT:ISTA:7680}, year = {2020}, }