Search

Background Depression is a heterogeneous mental illness. Neurostimulation treatments, by targeting specific nodes within the brain's emotion-regulation network, may be useful both as therapies and as probes for identifying clinically relevant depression subtypes. Methods Here, we applied 20 sessions of magnetic resonance imaging-guided repetitive transcranial magnetic stimulation (rTMS) to the dorsomedial prefrontal cortex in 47 unipolar or bipolar patients with a medication-resistant major depressive episode. Results Treatment response was strongly bimodal, with individual patients showing either minimal or marked improvement. Compared with responders, nonresponders showed markedly higher baseline anhedonia symptomatology (including pessimism, loss of pleasure, and loss of interest in previously enjoyed activities) on item-by-item examination of Beck Depression Inventory-II and Quick Inventory of Depressive Symptomatology ratings. Congruently, on baseline functional magnetic resonance imaging, nonresponders showed significantly lower connectivity through a classical reward pathway comprising ventral tegmental area, striatum, and a region in ventromedial prefrontal cortex. Responders and nonresponders also showed opposite patterns of hemispheric lateralization in the connectivity of dorsomedial and dorsolateral regions to this same ventromedial region. Conclusions The results suggest distinct depression subtypes, one with preserved hedonic function and responsive to dorsomedial rTMS and another with disrupted hedonic function, abnormally lateralized connectivity through ventromedial prefrontal cortex, and unresponsive to dorsomedial rTMS. Future research directly comparing the effects of rTMS at different targets, guided by neuroimaging and clinical presentation, may clarify whether hedonia/reward circuit integrity is a reliable marker for optimizing rTMS target selection.

A set of proteins is a complex system whose elements are interrelated on the concept of sequence- and structure-based similarity. Here, we applied a similarity network-based methodology for the representation and analysis of protein sequences and structures sets using a non-redundant set of 311 proteins and three different information criteria based on sequence-derived features, sequence local alignment and structural alignment. A wide set of measurements, like network degree, clustering coefficient, characteristic path length and vertex centrality were utilized to characterize the networks’ topology. Protein similarity networks were found medium or highly interconnected and the existence of both clusters and random edges classified their fully connected versions as Small World Networks (SWNs). The SWN architecture was able to host the continuous similarity transition among proteins and model the protein information flow during evolution. Recently reported ancestral elements, like the α/β class and certain folds, were remarkably found to act as hubs in the networks. Additionally, the moderate information value of sequence-derived features when used for fold and class assignment was shown on a network basis. The methodology described here can be applied for the analysis of other complex systems which consist of interrelated elements and a certain information flow.

We prove that Consistency with Certainty, Continuity and Extension assure that Betweenness and Substitution-independence together are equivalent to Weak Substitution. In this way we get an alternative axiomatic approach to the MαΦ theory whose intuitive appeal is greater then Chew's one.

The final publication is available at Springer via http://dx.doi.org/10.1007/s13278-016-0346-9. We rethink and extend the concepts of position and role in a network, basing them on various well-known measures that were not previously associated with these concepts, such as geodesic distance and modularity. The effectiveness of our new role and position detection algorithms is evaluated both qualitatively and quantitatively, on synthetic and real data, showing that we can identify new types of meaningful patterns in networks.