Search

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.

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.

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

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.

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.

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.