Hyperparameter on-line learning of stochastic resonance based threshold networks

Aiming at training the feed-forward threshold neural network consisting of nondifferentiable activation functions, the approach of noise injection forms a stochastic resonance based threshold network that can be optimized by various gradient-based optimizers. The introduction of injected noise extends the noise level into the parameter space of the designed threshold network, but leads to a highly non-convex optimization landscape of the loss function. Thus, the hyperparameter on-line learning procedure with respective to network weights and noise levels becomes of challenge. It is shown that the Adam optimizer, as an adaptive variant of stochastic gradient descent, manifests its superior learning ability in training the stochastic resonance based threshold network effectively. Experimental results demonstrate the significant improvement of performance of the designed threshold network trained by the Adam optimizer for function approximation and image classification.     

Explosive synchronization in a mobile network in the presence of a positive feedback mechanism

Synchronization is a process that describes the coherent dynamics of a large ensemble of interacting units.The study of explosive synchronization transition attracts considerable attention.Here,I report the explosive transition within the framework of a mobile network,while each oscillator is controlled by global-order parameters of the system.Using numerical simulation,I find that the explosive synchronization(ES)transition behavior can be controlled by simply adjusting the fraction of controlled oscillators.The influences of some parameters on explosive synchronization are studied.Moreover,due to the presence of the positive feedback mechanism,I prevent the occurrence of the synchronization of continuous-phase transition and make phase transition of the system a first-order phase transition accompanied by a hysteresis loop.     

Pattern transition and regulation in a subthalamopallidal network under electromagnetic effect

Although the significant roles of magnetic induction and electromagnetic radiation in the neural system have been widely studied, their influence on Parkinson's disease (PD) has yet to be well explored. By virtue of the magnetic flux variable, this paper studies the transition of firing patterns induced by magnetic induction and the regulation effect of external magnetic radiation on the firing activities of the subthalamopallidal network in basal ganglia. We find: (i) The network reproduces five typical waveforms corresponding to the severity of symptoms: weak cluster, episodic, continuous cluster, episodic, and continuous wave. (ii) Magnetic induction is a double-edged sword for the treatment of PD. Although the increase of magnetic coefficient may lead the physiological firing activity to transfer to pathological firing activity, it also can regulate the pathological intensity firing activity with excessive β-band power transferring to the physiological firing pattern with weak β-band power. (iii) External magnetic radiation could inhibit continuous tremulous firing and β-band power of subthalamic nucleus (STN), which means the severity of symptoms weakened. Especially, the bi-parameter plane of the regulation region shows that a short pulse period of magnetic radiation and a medium level of pulse percentage can well regulate pathological oscillation. This work helps to understand the firing activity of the subthalamopallidal network under electromagnetic effect. It may also provide insights into the mechanisms behind the electromagnetic therapy of PD-related firing activity.     

The Special “Morse Code” between Solvent Polarity and Morphology Flexibility in a POSS–Dendrimer Supramolecular System

An amazing phenomenon of the relative magnitude of modulus of two liquid-crystal (LC) gels is found inverted under/above their phase transition temperature T_LC-iso, which is further proved to be caused by their diverse morphology flexibility. By testing the polarity of two LCs, gelator POSS-G1-Boc (POSS=polyhedral oligomeric silsesquioxane) was discovered to self-assemble into more flexible structures in a relatively low polar LC, whereas more rigid ones are formed in higher polar LC. Hence, a fitting function to connect morphology flexibility with solvent polarity was established, which can even be generalized to a number of common solvents. Experimental observations and coarse-grained molecular dynamics simulations revealed that solvent polarity mirrors a “Morse code”, with each “code” corresponding to a specific morphology flexibility.     

PtII-Catalyzed Hydroxylation of Terminal Aliphatic C(sp3)−H Bonds with Molecular Oxygen

The practical application of Shilov-type Pt catalysis to the selective hydroxylation of terminal aliphatic C−H bonds remains a formidable challenge, due to difficulties in replacing Pt^IV with a more economically viable oxidant, particularly O₂. We report the potential of employing FeCl₂ as a suitable redox mediator to overcome the kinetic hurdles related to the direct use of O₂ in the Pt reoxidation. For the selective conversion of butyric acid to γ-hydroxybutyric acid (GHB), a significantly enhanced catalyst activity and stability (turnover numbers (TON)>30) were achieved under 20 bar O₂ in comparison to current state-of-the-art systems (TON<10). In this regard, essential reaction parameters affecting the overall activity were identified, along with specific additives to attain catalyst stability at longer reaction times. Notably, deactivation by reduction to Pt⁰ was prevented by the addition of monodentate pyridine derivatives, such as 2-fluoropyridine, but also by introducing varying partial pressures of N₂ in the gaseous atmosphere. Finally, stability tests revealed the involvement of Pt^II and FeCl₂ in catalyzing the non-selective overoxidation of GHB. Accordingly, in situ esterification with boric acid proved to be a suitable strategy to maintain enhanced selectivities at much higher conversions (TON>60). Altogether, a useful catalytic system for the selective hydroxylation of primary aliphatic C−H bonds with O₂ is presented.     

Design and Synthesis of Imidazole and Triazole Pyrazoles as Mycobacterium Tuberculosis CYP121A1 Inhibitors

The emergence of untreatable drug-resistant strains of Mycobacterium tuberculosis is a major public health problem worldwide, and the identification of new efficient treatments is urgently needed. Mycobacterium tuberculosis cytochrome P450 CYP121A1 is a promising drug target for the treatment of tuberculosis owing to its essential role in mycobacterial growth. Using a rational approach, which includes molecular modelling studies, three series of azole pyrazole derivatives were designed through two synthetic pathways. The synthesized compounds were biologically evaluated for their inhibitory activity towards M. tuberculosis and their protein binding affinity (K_D). Series 3 biarylpyrazole imidazole derivatives were the most effective with the isobutyl ( 10 f ) and tert-butyl ( 10 g ) compounds displaying optimal activity (MIC 1.562 μg/mL, K_D 0.22 μM ( 10 f ) and 4.81 μM ( 10 g )). The spectroscopic data showed that all the synthesised compounds produced a type II red shift of the heme Soret band indicating either direct binding to heme iron or (where less extensive Soret shifts are observed) putative indirect binding via an interstitial water molecule. Evaluation of biological and physicochemical properties identified the following as requirements for activity: LogP >4, H-bond acceptors/H-bond donors 4/0, number of rotatable bonds 5–6, molecular volume >340 ų, topological polar surface area <40 Ų.     

Green Synthesis of new Trizole Based Heterocyclic Amino Acids Ligands and their Transition Metal Complexes. Characterization,Kinetics, Antimicrobial and Docking Studies

Two novel amino acids imine ligands (H₂L₁ and H₂L₂) have been synthesized using green condensation reaction from 2‐[3‐Amino‐5‐(2‐hydroxy‐phenyl)‐5‐methyl‐1,5‐dihydro‐[1, 2, 4]triazol‐4‐yl]‐3‐(1H‐indol‐3‐yl)‐propionic acid with benzaldehyde/p‐flouro benzaldehyde (1:1 molar ratio) in the presence of lemon juice as a natural acidic catalyst in aqueous medium. Their transition metal complexes have been prepared in a molar ratio (1:1). Characterization of the ligands and complexes using elemental analysis, spectroscopic studies, ¹HNMR, ¹³CNMR, and thermal analysis has been reported. E*, ΔH*, ΔS* and ΔG* thermodynamic parameters, were calculated to throw more light on the nature of changes accompanying the thermal decomposition process of these complexes. The molar conductance measurement of metal complexes showed nonelectrolyte behavior. The metal complexes of the two ligands have tetrahedral geometry with a general molecular structure [M(H₂L)X_n], where [(M = Mn (II), Co (II), Cu (II) and Zn (II), X = Cl, n = 2]; M = VO (II), X = SO₄, n = 1] for H₂L₁. [M = Co (II), Cu (II), Zn (II)] for H₂L₂. Antibacterial activity of the complexes against (Bacillis subtilis, Micrococcus luteus, Escherichia coli), also antifungal activity against (Aspergillus niger, Candida Glabarta, Saccharomyces cerevisiae) have been screened. The results showed that all complexes have antimicrobial activity higher than free ligands. Molecular docking studies results showed that, all the synthesized compounds having minimum binding energy and have good affinity toward the active pocket, thus, they may be considered as good inhibitor of targeting PDB code: 1SC7 (Human DNA Topo‐isomerase I).     

A Bait-and-Switch Method for the Construction of Artificial Esterases for Substrate-Selective Hydrolysis

Outcomes of chemical reactions are generally dominated by the intrinsic reactivities of reaction partners, but enzymes frequently override such constraints to transform less reactive molecules in the presence of more reactive ones. Despite the attractiveness of such catalysis, it is difficult to build synthetic catalysts with these features. Micellar imprinting is a powerful method to create template-complementary binding sites inside protein-sized water-soluble nanoparticles. When a photocleavable functional monomer was used to bind two phosphonate/phosphate templates as transition-state analogues, active sites with predetermined size and shape were formed inside doubly cross-linked micelles through molecular imprinting. Postmodification replaced the binding group with a catalytic pyridyl group, forming highly selective artificial esterases. The catalysts displayed enzyme-like kinetics and turnover numbers that were in the hundreds. The selectivity of the catalysts, derived from the substrate-complementary imprinted active sites, enabled transformation of less reactive esters in the presence of more reactive ones.     

A Self-Assembled Cage with Endohedral Acid Groups both Catalyzes Substitution Reactions and Controls Their Molecularity

A self-assembled Fe₄L₆ cage complex internally decorated with acid functions is capable of accelerating the thioetherification of activated alcohols, ethers and amines by up to 1000-fold. No product inhibition is seen, and effective supramolecular catalysis can occur with as little as 5 % cage. The substrates are bound in the host with up to micromolar affinities, whereas the products show binding that is an order of magnitude weaker. Most importantly, the cage host alters the molecularity of the reaction: whereas the reaction catalyzed by simple acids is a unimolecular, S_N1-type substitution process, the rate of the host-mediated process is dependent on the concentration of nucleophile. The molecularity of the cage-catalyzed reaction is substrate-dependent, and can be up to bimolecular. In addition, the catalysis can be prevented by a large excess of nucleophile, where substrate inhibition dominates, and the use of tritylated anilines as substrates causes a negative feedback loop, whereby the liberated product destroys the catalyst and stops the reaction.     

Code2vect: An efficient heterogenous data classifier and nonlinear regression technique

The aim of this paper is to present a new classification and regression algorithm based on Artificial Intelligence. The main feature of this algorithm, which will be called Code2Vect, is the nature of the data to treat: qualitative or quantitative and continuous or discrete. Contrary to other artificial intelligence techniques based on the “Big-Data,” this new approach will enable working with a reduced amount of data, within the so-called “Smart Data” paradigm. Moreover, the main purpose of this algorithm is to enable the representation of high-dimensional data and more specifically grouping and visualizing this data according to a given target. For that purpose, the data will be projected into a vectorial space equipped with an appropriate metric, able to group data according to their affinity (with respect to a given output of interest). Furthermore, another application of this algorithm lies on its prediction capability. As it occurs with most common data-mining techniques such as regression trees, by giving an input the output will be inferred, in this case considering the nature of the data formerly described. In order to illustrate its potentialities, two different applications will be addressed, one concerning the representation of high-dimensional and categorical data and another featuring the prediction capabilities of the algorithm.