Functional Kernel Density Estimation: Point and Fourier Approaches to Time Series Anomaly Detection.

We present an unsupervised method to detect anomalous time series among a collection of time series. To do so, we extend traditional Kernel Density Estimation for estimating probability distributions in Euclidean space to Hilbert spaces. The estimated probability densities we derive can be obtained formally through treating each series as a point in a Hilbert space, placing a kernel at those points, and summing the kernels (a "point approach"), or through using Kernel Density Estimation to approximate the distributions of Fourier mode coefficients to infer a probability density (a "Fourier approach"). We refer to these approaches as Functional Kernel Density Estimation for Anomaly Detection as they both yield functionals that can score a time series for how anomalous it is. Both methods naturally handle missing data and apply to a variety of settings, performing well when compared with an outlyingness score derived from a boxplot method for functional data, with a Principal Component Analysis approach for functional data, and with the Functional Isolation Forest method. We illustrate the use of the proposed methods with aviation safety report data from the International Air Transport Association (IATA).

Phase Behavior of Poly(ethylene oxide) in Room Temperature Ionic Liquids: A Molecular Simulation and Deep Neural Network Study.

The phase behavior of polymers in room temperature ionic liquids is a topic of considerable interest. In this work we study the phase diagram of poly(ethylene oxide) in four imidazolium ionic liquids (ILs) using molecular simulation. We develop united atom models for 1-butyl-2,3-dimethylimidazolium ([BMMIM]), 1-ethyl-2,3-dimethylimidazolium ([EMMIM]), and 1-ethyl-3-methylimidazolium ([EMIM]) in an analogous fashion to previously developed models for 1-butyl-3-methylimidazolium ([BMIM]) and tetrafluoroborate ([BF4]) using symmetry-adapted perturbation theory. At high temperatures we obtain the coexistence concentrations using an interface method where the polymer and IL are simulated in a large elongated box, and an interface between coexisting phases is formed. At lower temperatures we use a deep neural network (DNN) method. The input descriptors for the DNN are the cohesive energy of mixing, the volume change of mixing, and the coordination numbers between cation and polymer, all of which are obtained from simulations of mixed systems at a series of temperatures. The DNN is trained by using the phase-separated systems at high temperatures and a mixed phase at low temperatures. The method predicts a lower critical solution temperature which decreases as the alkyl chain length on the cation is decreased, consistent with experiment. The simulations show that methylation of the cation has little effect on the phase diagram. This is in contrast to what is seen in experiments but could be because the polymer chains in the simulations are too short. At low temperatures the chains display two conformational motifs, namely a crown ether conformation and a ring conformation, each of which can wrap the chain around a single cation. This provides the entropic penalty for mixing and a reason for demixing as the temperature is raised. Such conformations might not be possible for longer chains. The combination of data-driven techniques and molecular simulation shows promise in the study of the phase behavior and physical properties of complex fluids.

Phase behavior of continuous-space systems: A supervised machine learning approach.

The phase behavior of complex fluids is a challenging problem for molecular simulations. Supervised machine learning (ML) methods have shown potential for identifying the phase boundaries of lattice models. In this work, we extend these ML methods to continuous-space systems. We propose a convolutional neural network model that utilizes grid-interpolated coordinates of molecules as input data of ML and optimizes the search for phase transitions with different filter sizes. We test the method for the phase diagram of two off-lattice models, namely, the Widom-Rowlinson model and a symmetric freely jointed polymer blend, for which results are available from standard molecular simulations techniques. The ML results show good agreement with results of previous simulation studies with the added advantage that there is no critical slowing down. We find that understanding intermediate structures near a phase transition and including them in the training set is important to obtain the phase boundary near the critical point. The method is quite general and easy to implement and could find wide application to study the phase behavior of complex fluids.

A simulation method for the phase diagram of complex fluid mixtures.

The phase behavior of complex fluid mixtures is of continuing interest, but obtaining the phase diagram from computer simulations can be challenging. In the Gibbs ensemble method, for example, each of the coexisting phases is simulated in a different cell, and ensuring the equality of chemical potentials of all components requires the transfer of molecules from one cell to the other. For complex fluids such as polymers, successful insertions are rare. An alternative method is to simulate both coexisting phases in a single simulation cell, with an interface between them. The challenge here is that the interface position moves during the simulation, making it difficult to determine the concentration profile and coexisting concentrations. In this work, we propose a new method for single cell simulations that uses a spatial concentration autocorrelation function to (spatially) align instantaneous concentration profiles from different snapshots. This allows one to obtain average concentration profiles and hence the coexisting concentrations. We test the method by calculating the phase diagrams of two systems: the Widom-Rowlinson model and the symmetric blends of freely jointed polymer molecules for which phase diagrams from conventional methods are available. Excellent agreement is found, except in the neighborhood of the critical point where the interface is broad and finite size effects are important. The method is easy to implement and readily applied to any mixture of complex fluids.

Three-dimensional bioprinting of multilayered constructs containing human mesenchymal stromal cells for osteochondral tissue regeneration in the rabbit knee joint.

The use of cell-rich hydrogels for three-dimensional (3D) cell culture has shown great potential for a variety of biomedical applications. However, the fabrication of appropriate constructs has been challenging. In this study, we describe a 3D printing process for the preparation of a multilayered 3D construct containing human mesenchymal stromal cells with a hydrogel comprised of atelocollagen and supramolecular hyaluronic acid (HA). This construct showed outstanding regenerative ability for the reconstruction of an osteochondral tissue in the knee joints of rabbits. We found that the use of a mechanically stable, host-guest chemistry-based hydrogel was essential and allowed two different types of extracellular matrix (ECM) hydrogels to be easily printed and stacked into one multilayered construct without requiring the use of potentially harmful chemical reagents or physical stimuli for post-crosslinking. To the best of our knowledge, this is the first study to validate the potential of a 3D printed multilayered construct consisting of two different ECM materials (atelocollagen and HA) for heterogeneous tissue regeneration using an in vivo animal model. We believe that this 3D printing-based platform technology can be effectively exploited for regeneration of various heterogeneous tissues as well as osteochondral tissue.

Genetically engineered mesenchymal stem cell therapy using self-assembling supramolecular hydrogels.

Stem cell therapy has attracted a great deal of attention for treating intractable diseases such as cancer, stroke, liver cirrhosis, and ischemia. Especially, mesenchymal stem cells (MSCs) have been widely investigated for therapeutic applications due to the advantageous characteristics of long life-span, facile isolation, rapid proliferation, prolonged transgene expression, hypo-immunogenicity, and tumor tropism. MSCs can exert their therapeutic effects by releasing stress-induced therapeutic molecules after their rapid migration to damaged tissues. Recently, to improve the therapeutic efficacy, genetically engineered MSCs have been developed for therapeutic transgene expression by viral gene transduction and non-viral gene transfection. In general, the number of therapeutic cells for injection should be more than several millions for effective cell therapy. Adequate carriers for the controlled delivery of MSCs can reduce the required cell numbers and extend the duration of therapeutic effect, which provide great benefits for chronic disease patients. In this review, we describe genetic engineering of MSCs, recent progress of self-assembling supramolecular hydrogels, and their applications to cell therapy for intractable diseases and tissue regeneration.

Supramolecular hydrogels for long-term bioengineered stem cell therapy.

Synthetic hydrogels have been extensively investigated as artificial extracellular matrices (ECMs) for tissue engineering in vitro and in vivo. Crucial challenges for such hydrogels are sustaining long-term cytocompatible encapsulation and providing appropriate cues at the right place and time for spatio-temporal control of the cells. Here, in situ supramolecularly assembled and modularly modified hydrogels for long-term engineered mesenchymal stem cell (eMSC) therapy are reported using cucurbit[6]uril-conjugated hyaluronic acid (CB[6]-HA), diaminohexane conjugated HA (DAH-HA), and drug-conjugated CB[6] (drug-CB[6]). The eMSCs producing enhanced green fluorescence protein (EGFP) remain alive and emit the fluorescence within CB[6]/DAH-HA hydrogels in mice for more than 60 d. Furthermore, the long-term expression of mutant interleukin-12 (IL-12M) by eMSCs within the supramolecular hydrogels results in effective inhibition of tumor growth with a significantly enhanced survival rate. Taken together, these findings confirm the feasibility of supramolecular HA hydrogels as 3D artificial ECMs for cell therapies and tissue engineering applications.

3D tissue engineered supramolecular hydrogels for controlled chondrogenesis of human mesenchymal stem cells.

Despite a wide investigation of hydrogels as an artificial extracellular matrix, there are few scaffold systems for the facile spatiotemporal control of mesenchymal stem cells (MSCs). Here, we report 3D tissue engineered supramolecular hydrogels prepared with highly water-soluble monofunctionalized cucurbit[6]uril-hyaluronic acid (CB[6]-HA), diaminohexane conjugated HA (DAH-HA), and drug conjugated CB[6] (drug-CB[6]) for the controlled chondrogenesis of human mesenchymal stem cells (hMSCs). The mechanical property of supramolecular HA hydrogels was modulated by changing the cross-linking density for the spatial control of hMSCs. In addition, the differentiation of hMSCs was temporally controlled by changing the release profiles of transforming growth factor-ß3 (TGF-ß3) and/or dexamethasone (Dexa) from the hydrolyzable Dexa-CB[6]. The effective chondrogenic differentiation of hMSCs encapsulated in the monoCB[6]/DAH-HA hydrogel with TGF-ß3 and Dexa-CB[6] was confirmed by biochemical glycosaminoglycan content analysis, real-time quantitative PCR, histological, and immunohistochemical analyses. Taken together, we could confirm the feasibility of cytocompatible monoCB[6]/DAH-HA hydrogels as a platform scaffold with controlled drug delivery for cartilage regeneration and other various tissue engineering applications.

In situ supramolecular assembly and modular modification of hyaluronic acid hydrogels for 3D cellular engineering.

A facile in situ supramolecular assembly and modular modification of biocompatible hydrogels were demonstrated using cucurbit[6]uril-conjugated hyaluronic acid (CB[6]-HA), diaminohexane-conjugated HA (DAH-HA), and tags-CB[6] for cellular engineering applications. The strong and selective host-guest interaction between CB[6] and DAH made possible the supramolecular assembly of CB[6]/DAH-HA hydrogels in the presence of cells. Then, the 3D environment of CB[6]/DAH-HA hydrogels was modularly modified by the simple treatment with various multifunctional tags-CB[6]. Furthermore, we could confirm in situ formation of CB[6]/DAH-HA hydrogels under the skin of nude mice by sequential subcutaneous injections of CB[6]-HA and DAH-HA solutions. The fluorescence of modularly modified fluorescein isothiocyanate (FITC)-CB[6] in the hydrogels was maintained for up to 11 days, reflecting the feasibility to deliver the proper cues for cellular proliferation and differentiation in the body. Taken together, CB[6]/DAH-HA hydrogels might be successfully exploited as a 3D artificial extracellular matrix for various tissue engineering applications.

Target specific hyaluronic acid-interferon alpha conjugate for the treatment of hepatitis C virus infection.

Interferon alpha (IFNα) conjugated with polyethylene glycol (PEG) has been widely used for the treatment of hepatitis C virus (HCV) infection as a once-a-week injection formulation. However, the PEGylated IFNα has a low efficacy of ca. 39% and a side effect after repeated injections possibly due to the non-specific delivery with PEGylation. In this work, target specific long-acting hyaluronic acid-interferon alpha (HA-IFNα) conjugate was successfully developed for the treatment of HCV infection. HA-IFNα conjugate was synthesized by coupling reaction between aldehyde modified HA and the N-terminal group of IFNα. The IFNα content could be controlled in the range of 2-9 molecules per single HA chain with a bioconjugation efficiency higher than 95%. According to in vitro anti-proliferation assay using Daudi cells, HA-IFNα conjugate showed a comparable biological activity to PEG-Intron. In vivo real-time bioimaging confirmed the target specific delivery of near-infrared fluorescence (NIRF) dye labeled HA-IFNα conjugate to the liver in mice. In addition, pharmacokinetic analysis revealed the enhanced residence time longer than 4 days. After tail-vein injection, HA-IFNα conjugate induced ca. 60% higher expression of 2',5'-oligoadenylate synthetase 1 (OAS 1) for innate immune responses to viral infection in the murine liver tissues than IFNα and PEG-Intron.

Theranostic systems assembled in situ on demand by host-guest chemistry.

Theranostic systems have been explored extensively for a diagnostic therapy in the forms of polymer conjugates, implantable devices, and inorganic nanoparticles. In this work, we report theranostic systems in situ assembled by host-guest chemistry responding to a request. As a model theranostic system on demand, cucurbit[6]uril-conjugated hyaluronate (CB[6]-HA) was synthesized and decorated with FITC-spermidine (spmd) and/or formyl peptide receptor like 1 (FPRL1) specific peptide-spmd by simple mixing in aqueous solution. The resulting (FITC-spmd and/or peptide-spmd)@CB[6]-HA was successfully applied to the bioimaging of its target-specific delivery to B16F1 cells with HA receptors and its therapeutic signal transduction with elevated Ca(2+) and phosphor-extracellular signal-regulated kinase (pERK) levels in FPRL1-expressing human breast adenocarcinoma (FPRL1/MCF-7) cells. Finally, we could confirm in vitro and in vivo stability of the highly specific host-guest interaction. The on-demand theranostic platform technology using host-guest chemistry can be exploited for various bioimaging, biosensing, drug delivery, and tissue engineering applications.

Supramolecular fishing for plasma membrane proteins using an ultrastable synthetic host-guest binding pair.

Membrane proteomics, the large-scale global analysis of membrane proteins, is often constrained by the efficiency of separating and extracting membrane proteins. Recent approaches involve conjugating membrane proteins with the small molecule biotin and using the receptor streptavidin to extract the labelled proteins. Despite the many advantages of this method, several shortcomings remain, including potential contamination by endogenously biotinylated molecules and interference by streptavidin during analytical stages. Here, we report a supramolecular fishing method for membrane proteins using the synthetic receptor-ligand pair cucurbit[7]uril-1-trimethylammoniomethylferrocene (CB[7]-AFc). CB[7]-conjugated beads selectively capture AFc-labelled proteins from heterogeneous protein mixtures, and AFc-labelling of cells results in the efficient capture of membrane proteins by these beads. The captured proteins can be recovered easily at room temperature by treatment with a strong competitor such as 1,1'-bis(trimethylammoniomethyl)ferrocene. This synthetic but biocompatible host-guest system may be a useful alternative to streptavidin-biotin for membrane proteomics as well as other biological and biotechnological applications.

Reduction-sensitive, robust vesicles with a non-covalently modifiable surface as a multifunctional drug-delivery platform.

The design and synthesis of a novel reduction-sensitive, robust, and biocompatible vesicle (SSCB[6]VC) are reported, which is self-assembled from an amphiphilic cucurbit[6]uril (CB[6]) derivative that contains disulfide bonds between hexaethylene glycol units and a CB[6] core. The remarkable features of SSCB[6]VC include: 1) facile, non-destructive, non-covalent, and modular surface modification using exceptionally strong host-guest chemistry; 2) high structural stability; 3) facile internalization into targeted cells by receptor-mediated endocytosis, and 4) efficient triggered release of entrapped drugs in a reducing environment such as cytoplasm. Furthermore, a significantly increased cytotoxicity of the anticancer drug doxorubicin to cancer cells is demonstrated using doxorubicin-loaded SSCB[6]VC, the surface of which is decorated with functional moieties such as a folate-spermidine conjugate and fluorescein isothiocyanate-spermidine conjugate as targeting ligand and fluorescence imaging probe, respectively. SSCB[6]VC with such unique features can be used as a highly versatile multifunctional platform for targeted drug delivery, which may find useful applications in cancer therapy. This novel strategy based on supramolecular chemistry and the unique properties of CB[6] can be extended to design smart multifunctional materials for biomedical applications including gene delivery.

Cucurbituril-based nanoparticles: a new efficient vehicle for targeted intracellular delivery of hydrophobic drugs.

Cucurbituril-based nanoparticles (CB[6]NPs) serve as new efficient vehicles for delivery of hydrophobic drugs, which have unique features including (1) a high drug loading capacity and efficiency, (2) noncovalently tunable surfaces, (3) efficient delivery of hydrophobic drugs into a cancer cell by receptor-mediated endocytosis, and (4) facile release of drugs into cytoplasm, which enhances the pharmaceutical effects of the drugs.

Cdk5 phosphorylates PLD2 to mediate EGF-dependent insulin secretion.

Insulin secretion from pancreatic beta-cells is an important process that affects the regulation of glucose level in the blood. In our previous study, we suggested that epidermal growth factor (EGF) stimulates insulin secretion by activating phospholipase D2 (PLD2) [H.Y. Lee, K. Yea, J. Kim, B.D. Lee, Y.C. Chae, H.S. Kim, D.W. Lee, S.H. Kim, J.H. Cho, C.J. Jin, D.S. Koh, K.S. Park, P.G. Suh, S.H. Ryu, 2007. Epidermal Growth Factor Increases Insulin Secretion and Lowers Blood Glucose in Diabetic Mice. J. Cell. Mol. Med. 5:5]. However, the specific mechanism by which PLD2 activation leads to insulin secretion was not determined. In this study, we suggest that the phosphorylation and activation of PLD2 by cyclin-dependent kinase 5 (Cdk5) is critical for EGF-dependent insulin secretion. We found that a Cdk5 inhibitor, roscovitine, and dominant-negative Cdk5 inhibited EGF-dependent PLD2 activation and insulin secretion. EGF stimulation activated Cdk5 activity in rat insulinoma RINm5F cells, and PLD2 phosphorylation by Cdk5 was observed in vitro and in cells. The kinetics of PLD2 phosphorylation correlates with the interaction between PLD2 and Cdk5 and its effect on EGF signaling. We determined that the phosphorylation site of PLD2 was located at Ser(134). PLD2-S134A did not show EGF-dependent phosphorylation and activation by Cdk5. Furthermore, this mutant was unable to mediate EGF-dependent insulin secretion in pancreatic beta cell lines, suggesting that the phosphorylation of PLD2 at Ser(134) by Cdk5 is critical for this process. The study results suggest that PLD2 is a new substrate of Cdk5 and that the phosphorylation of PLD2 by Cdk5 is involved in EGF-dependent insulin secretion.