Bias reduction for semi-competing risks frailty model with rare events: application to a chronic kidney disease cohort study in South Korea.

In a semi-competing risks model in which a terminal event censors a non-terminal event but not vice versa, the conventional method can predict clinical outcomes by maximizing likelihood estimation. However, this method can produce unreliable or biased estimators when the number of events in the datasets is small. Specifically, parameter estimates may converge to infinity, or their standard errors can be very large. Moreover, terminal and non-terminal event times may be correlated, which can account for the frailty term. Here, we adapt the penalized likelihood with Firth's correction method for gamma frailty models with semi-competing risks data to reduce the bias caused by rare events. The proposed method is evaluated in terms of relative bias, mean squared error, standard error, and standard deviation compared to the conventional methods through simulation studies. The results of the proposed method are stable and robust even when data contain only a few events with the misspecification of the baseline hazard function. We also illustrate a real example with a multi-centre, patient-based cohort study to identify risk factors for chronic kidney disease progression or adverse clinical outcomes. This study will provide a better understanding of semi-competing risk data in which the number of specific diseases or events of interest is rare.

Multiple-Kernel Support Vector Machine for Predicting Internet Gaming Disorder Using Multimodal Fusion of PET, EEG, and Clinical Features.

Internet gaming disorder (IGD) has become an important social and psychiatric issue in recent years. To prevent IGD and provide the appropriate intervention, an accurate prediction method for identifying IGD is necessary. In this study, we investigated machine learning methods of multimodal neuroimaging data including Positron Emission Tomography (PET), Electroencephalography (EEG), and clinical features to enhance prediction accuracy. Unlike the conventional methods which usually concatenate all features into one feature vector, we adopted a multiple-kernel support vector machine (MK-SVM) to classify IGD. We compared the prediction performance of standard machine learning methods such as SVM, random forest, and boosting with the proposed method in patients with IGD (N = 28) and healthy controls (N = 24). We showed that the prediction accuracy of the optimal MK-SVM using three kinds of modalities was much higher than other conventional machine learning methods, with the highest accuracy being 86.5%, the sensitivity 89.3%, and the specificity 83.3%. Furthermore, we deduced that clinical variables had the highest contribution to the optimal IGD prediction model and that the other two modalities were also indispensable. We found that more efficient integration of multimodal data through kernel combination could contribute to better performance of the prediction model. This study is a novel attempt to integrate each method from different sources and suggests that integrating each method, such as self-administrated reports, PET, and EEG, improves the prediction of IGD.

Identification of Major Psychiatric Disorders From Resting-State Electroencephalography Using a Machine Learning Approach.

We aimed to develop a machine learning (ML) classifier to detect and compare major psychiatric disorders using electroencephalography (EEG). We retrospectively collected data from medical records, intelligence quotient (IQ) scores from psychological assessments, and quantitative EEG (QEEG) at resting-state assessments from 945 subjects [850 patients with major psychiatric disorders (six large-categorical and nine specific disorders) and 95 healthy controls (HCs)]. A combination of QEEG parameters including power spectrum density (PSD) and functional connectivity (FC) at frequency bands was used to establish models for the binary classification between patients with each disorder and HCs. The support vector machine, random forest, and elastic net ML methods were applied, and prediction performances were compared. The elastic net model with IQ adjustment showed the highest accuracy. The best feature combinations and classification accuracies for discrimination between patients and HCs with adjusted IQ were as follows: schizophrenia = alpha PSD, 93.83%; trauma and stress-related disorders = beta FC, 91.21%; anxiety disorders = whole band PSD, 91.03%; mood disorders = theta FC, 89.26%; addictive disorders = theta PSD, 85.66%; and obsessive-compulsive disorder = gamma FC, 74.52%. Our findings suggest that ML in EEG may predict major psychiatric disorders and provide an objective index of psychiatric disorders.

In silico prediction of the full United Nations Globally Harmonized System eye irritation categories of liquid chemicals by IATA-like bottom-up approach of random forest method.

As an alternative to in vivo Draize rabbit eye irritation test, this study aimed to construct an in silico model to predict the complete United Nations (UN) Globally Harmonized System (GHS) for classification and labeling of chemicals for eye irritation category [eye damage (Category 1), irritating to eye (Category 2) and nonirritating (No category)] of liquid chemicals with Integrated approaches to testing and assessment (IATA)-like two-stage random forest approach. Liquid chemicals (n = 219) with 34 physicochemical descriptors and quality in vivo data were collected with no missing values. Seven machine learning algorithms (Naive Bayes, Logistic Regression, First Large Margin, Neural Net, Random Forest (RF), Gradient Boosted Tree, and Support Vector Machine) were examined for the ternary categorization of eye irritation potential at a single run through 10-fold cross-validation. RF, which performed best, was further improved by applying the 'Bottom-up approach' concept of IATA, namely, separating No category first, and discriminating Category 1 from 2, thereafter. The best performing training dataset achieved an overall accuracy of 73% and the correct prediction for Category 1, 2, and No category was 80%, 50%, and 77%, respectively for the test dataset. This prediction model was further validated with an external dataset of 28 chemicals, for which an overall accuracy of 71% was achieved.

Associations of personality and clinical characteristics with excessive Internet and smartphone use in adolescents: A structural equation modeling approach.

Adolescent Internet addiction is an important social issue entailing extensive use of Internet and smartphones and its side effects. This study identified relevant psychological factors that affect excessive Internet use (EIU) and excessive smartphone use (ESU) in adolescents using structural equation modeling (SEM). A sample of 714 individuals drawn from lists of middle school students in South Korea completed self-administered questionnaires, including Young's Internet Addiction Test (Y-IAT), the Smartphone Addiction Scale (SAS), and various clinical and psychological scales measuring depression, anxiety, attention deficit/hyperactivity disorder (ADHD), aggression, expression of anger, and the behavioral inhibition system (BIS)/activation system (BAS). The final model, fitted using SEM, showed that both clinical characteristics, including ADHD symptoms, aggression, expression of anger, depression, and anxiety, and personality characteristics, represented by BIS/BAS, played important roles in the severity of EIU or ESU. In particular, affective components such as depression and anxiety were significantly associated with both EIU and ESU, whereas aggression, the expression of anger, and ADHD symptoms affected only EIU. Furthermore, the association between ESU and EIU was significant. Although personality characteristics measured by the BIS and BAS scores did not have direct effects on addiction, they were associated with clinical features and might be risk factors for addiction. The model revealed significant pathways from personality and clinical features to EIU and ESU in adolescents and informed our basic understanding of the meaningful predictors of these addictions and their direct and indirect influences.

Correction: Prokaryotic soluble overexpression and purification of human VEGF165 by fusion to a maltose binding protein tag.

[This corrects the article DOI: 10.1371/journal.pone.0156296.].

Soluble Prokaryotic Expression and Purification of Human Interferon Alpha-2b Using a Maltose-Binding Protein Tag.

Human interferon alpha-2b (IFNα-2b) has therapeutic applications as an antiviral and antiproliferative drug and has been used for a wide range of indications. Efficient production of IFNα-2b in Escherichia coli has been difficult because the protein tends to form inclusion bodies. This obstacle has garnered interest in efficiently expressing IFNα-2b and overcoming its poor solubility. In this study, seven N-terminal fusion partners - hexahistidine (His6), thioredoxin, glutathione S-transferase (GST), maltose-binding protein (MBP), N-utilization substance protein A, protein disulfide bond isomerase (PDI), and b'a' domain of PDI - were tested for soluble overexpression of codon-optimized IFNα-2b in E. coli. Low temperature increased the expression level of all of the tagged proteins except for the GST fusion. All the tags, except for His6 and GST, improved solubility. We purified IFNα-2b from the MBP-tagged fusion using immobilized metal affinity chromatography and anion exchange chromatography, and obtained a final yield of 7.2 mg from an initial 500-ml culture. The endotoxin level was 0.46 EU/µg. Biological activity was demonstrated using a luciferase assay, which showed a dose-dependent response with a calculated EC50 of 10.3 ± 5.9 pM. Our results demonstrate that using an MBP-tagged fusion is an efficient way to produce pure IFNα-2b.

Prokaryotic Soluble Overexpression and Purification of Human VEGF165 by Fusion to a Maltose Binding Protein Tag.

Human vascular endothelial growth factor (VEGF) is a key regulator of angiogenesis and plays a central role in the process of tumor growth and metastatic dissemination. Escherichia coli is one of the most common expression systems used for the production of recombinant proteins; however, expression of human VEGF in E. coli has proven difficult because the E. coli-expressed VEGF tends to be misfolded and forms inclusion bodies, resulting in poor solubility. In this study, we successfully produced semi-preparative amounts of soluble bioactive human VEGF165 (hVEGF). We created seven N-terminal fusion tag constructs with hexahistidine (His6), thioredoxin (Trx), glutathione S-transferase (GST), maltose-binding protein (MBP), N-utilization substance protein A (NusA), human protein disulfide isomerase (PDI), and the b'a' domain of PDI (PDIb'a'), and tested each construct for soluble overexpression in E. coli. We found that at 18°C, 92.8% of the MBP-tagged hVEGF to be soluble and that this tag significantly increased the protein's solubility. We successfully purified 0.8 mg of pure hVEGF per 500 mL cell culture. The purified hVEGF is stable after tag cleavage, contains very low levels of endotoxin, and is 97.6% pure. Using an Flk1+ mesodermal precursor cell (MPC) differentiation assay, we show that the purified hVEGF is not only bioactive but has similar bioactivity to hVEGF produced in mammalian cells. Previous reports on producing hVEGF in E. coli have all been based on refolding of the protein from inclusion bodies. To our knowledge, this is the first report on successfully expressing and purifying soluble hVEGF in E. coli.

Soluble overexpression and purification of bioactive human CCL2 in E. coli by maltose-binding protein.

Human chemokine (C-C motif) ligand 2 (hCCL2) is a small cytokine in the CC chemokine family that attracts monocytes, memory T lymphocytes, and natural killer cells to the site of tissue injury- or infection-induced inflammation. hCCL2 has been implicated in the pathogeneses of diseases characterized by monocytic infiltrates, including psoriasis, rheumatoid arthritis, atherosclerosis, multiple sclerosis, and insulin-resistant diabetes. The prokaryotic overexpression of hCCL2 has been investigated previously in an attempt to develop biomedical applications for this factor, but this has been hampered by protein misfolding and aggregation into inclusion bodies. In our present study, we screened 7 protein tags-Trx, GST, MBP, NusA, His8, PDI, and PDIb'a'-for their ability to allow the soluble overexpression of hCCL2. Three tags-MBP, His8, and PDI-solubilized more than half of the expressed hCCL2 fusion proteins. Lowering the expression temperature to 18 °C significantly further improved the solubility of all fusion proteins. MBP was chosen for further study based on its solubility, expression level, ease of purification, and tag size. MBP-CCL2 was purified using conventional chromatography and cleaved using TEV or Factor Xa proteases. Biological activity was assessed using luciferase and cell migration assays. Factor Xa-cleaved hCCL2 was found to be active and TEV-cleaved hCCL2 showed relatively less activity. This is probably because the additional glycine residues present at the N-terminus of hCCL2 following TEV digestion interfere with the binding of hCCL2 to its receptor.

Soluble prokaryotic expression and purification of crotamine using an N-terminal maltose-binding protein tag.

Crotamine is a peptide toxin found in the venom of the rattlesnake Crotalus durissus terrificus. Interestingly, crotamine demonstrates promising anticancer, antimicrobial, and antifungal activities. The crotamine peptide can also deliver plasmids into rapidly dividing cells, such as cancer and stem cells, and demonstrates potent analgesic effects. Efficiently producing crotamine in mammalian cells is difficult because it is both cell-permeable and cytotoxic. Prokaryotic expression of this peptide is also difficult to maintain because it does not fold properly in the cytoplasm, resulting in aggregation and in the formation of inclusion bodies. In our current study, we show for the first time that N-terminal fusion with three protein tags-N-utilization substance protein A (NusA), protein disulfide isomerase b'a' domain (PDIb'a'), and maltose-binding protein (MBP)-enables the soluble overexpression of crotamine in the cytoplasm of Escherichia coli. MBP-tagged crotamine was purified using Ni affinity, anion exchange, and MBP chromatography. The tag was cleaved using TEV protease, and the final product was pure on a silver-stained gels. In total, 0.9 mg pure crotamine was obtained from each liter of bacterial culture with endotoxin level approximately 0.15 EU/µg, which is low enough to use in biomedical applications. The identity and intramolecular disulfide bonds were confirmed using MALDI-TOF MS analysis. Purified crotamine inhibited the hKv1.3 channel (but not hKv1.5) in a dose-dependent manner with IC50 value of 67.2 ± 44.7 nM (n = 10), indicating the correct protein folding. The crotamine product fused with MBP at its N-terminus also inhibited the hKv1.3 channel, suggesting that the N-terminus is not involved in the channel binding of the toxin.

Prokaryotic soluble overexpression and purification of bioactive human growth hormone by fusion to thioredoxin, maltose binding protein, and protein disulfide isomerase.

Human growth hormone (hGH) is synthesized by somatotroph cells of the anterior pituitary gland and induces cell proliferation and growth. This protein has been approved for the treatment of various conditions, including hGH deficiency, chronic renal failure, and Turner syndrome. Efficient production of hGH in Escherichia coli (E. coli) has proven difficult because the E. coli-expressed hormone tends to aggregate and form inclusion bodies, resulting in poor solubility. In this study, seven N-terminal fusion partners, hexahistidine (His6), thioredoxin (Trx), glutathione S-transferase (GST), maltose-binding protein (MBP), N-utilization substance protein A (NusA), protein disulfide bond isomerase (PDI), and the b'a' domain of PDI (PDIb'a'), were tested for soluble overexpression of codon-optimized hGH in E. coli. We found that MBP and hPDI tags significantly increased the solubility of the hormone. In addition, lowering the expression temperature to 18°C also dramatically increased the solubility of all the fusion proteins. We purified hGH from MBP-, PDIb'a'-, or Trx-tagged hGH expressed at 18°C in E. coli using simple chromatographic techniques and compared the final purity, yield, and activity of hGH to assess the impact of each partner protein. Purified hGH was highly pure on silver-stained gel and contained very low levels of endotoxin. On average, ∼37 mg, ∼12 mg, and ∼7 mg of hGH were obtained from 500 mL-cell cultures of Trx-hGH, MBP-hGH, and PDIb'a'-hGH, respectively. Subsequently, hGH was analyzed using mass spectroscopy to confirm the presence of two intra-molecular disulfide bonds. The bioactivity of purified hGHs was demonstrated using Nb2-11 cell.

Soluble expression of human leukemia inhibitory factor with protein disulfide isomerase in Escherichia coli and its simple purification.

Human leukemia inhibitory factor (hLIF) is a multifunctional cytokine that is essential for maintaining the pluripotency of embryonic stem cells. hLIF may be also be useful in aiding fertility through its effects on increasing the implantation rate of fertilized eggs. Thus these applications in biomedical research and clinical medicine create a high demand for bioactive hLIF. However, production of active hLIF is problematic since eukaryotic cells demonstrate limited expression and prokaryotic cells produce insoluble protein. Here, we have adopted a hybrid protein disulfide isomerase design to increase the solubility of hLIF in Escherichia coli. Low temperature expression of hLIF fused to the b'a' domain of protein disulfide isomerase (PDIb'a') increased the soluble expression in comparison to controls. A simple purification protocol for bioactive hLIF was established that includes removal of the PDIb'a' domain by cleavage by TEV protease. The resulting hLIF, which contains one extra glycine residue at the N-terminus, was highly pure and demonstrated endotoxin levels below 0.05 EU/µg. The presence of an intramolecular disulfide bond was identified using mass spectroscopy. This purified hLIF effectively maintained the pluripotency of a murine embryonic stem cell line. Thus we have developed an effective method to produce a pure bioactive version of hLIF in E. coli for use in biomedical research.

Expression and purification of biologically active human FGF2 containing the b'a' domains of human PDI in Escherichia coli.

Among the members of the fibroblast growth factor (FGF) family that affect the growth, differentiation, migration, and survival of many cell types, FGF2 is the most abundant in the central nervous system. Because of its wound healing effects, FGF2 has potential as a therapeutic agent. The protein is also added to the culture media to maintain stem cells. Expression and purification procedures for FGF2 that are highly efficient and low cost have been intensively investigated for the past two decades. Our current study focuses on the purification of FGF2 fused with b'a' domains of human protein disulfide isomerase to elevate overexpression, solubility, and stability with a simplified experimental procedure using only ion exchange chromatography, as well as on the confirmation of the biological activity of FGF2 on fibroblast Balb/c 3T3 cells and hippocampal neural cells.

Temperature increase due to Joule heating in a nanostructured MgO-based magnetic tunnel junction over a wide current-pulse range.

The temperature increase due to Joule heating in a nanopillar of a magnetic tunnel junction sandwiched by top and bottom electrodes was calculated by the finite element method. The results for the critical condition for the current-induced magnetization switching measured over a wide current-pulse range were taken from the literature. At long pulse widths, the temperature increase was solely dependent on the magnitude of the critical current density. However, no saturation in the temperature increase occurred for short pulse widths. In this case, the temperature increase additionally depended on the pulse width, so that a broad maximum occurred in the pulse width (or the critical current density) dependence of the temperature increase. The original results for the critical condition were corrected by accounting for the temperature increase and these corrected results, together with the Slonczewski equation, were used to extract an accurate value for the thermal stability factor.