A deep learning framework for identifying essential proteins based on multiple biological information.

BACKGROUND: Essential Proteins are demonstrated to exert vital functions on cellular processes and are indispensable for the survival and reproduction of the organism. Traditional centrality methods perform poorly on complex protein-protein interaction (PPI) networks. Machine learning approaches based on high-throughput data lack the exploitation of the temporal and spatial dimensions of biological information. RESULTS: We put forward a deep learning framework to predict essential proteins by integrating features obtained from the PPI network, subcellular localization, and gene expression profiles. In our model, the node2vec method is applied to learn continuous feature representations for proteins in the PPI network, which capture the diversity of connectivity patterns in the network. The concept of depthwise separable convolution is employed on gene expression profiles to extract properties and observe the trends of gene expression over time under different experimental conditions. Subcellular localization information is mapped into a long one-dimensional vector to capture its characteristics. Additionally, we use a sampling method to mitigate the impact of imbalanced learning when training the model. With experiments carried out on the data of Saccharomyces cerevisiae, results show that our model outperforms traditional centrality methods and machine learning methods. Likewise, the comparative experiments have manifested that our process of various biological information is preferable. CONCLUSIONS: Our proposed deep learning framework effectively identifies essential proteins by integrating multiple biological data, proving a broader selection of subcellular localization information significantly improves the results of prediction and depthwise separable convolution implemented on gene expression profiles enhances the performance.

Evaluating metagenomics tools for genome binning with real metagenomic datasets and CAMI datasets.

BACKGROUND: Shotgun metagenomics based on untargeted sequencing can explore the taxonomic profile and the function of unknown microorganisms in samples, and complement the shortage of amplicon sequencing. Binning assembled sequences into individual groups, which represent microbial genomes, is the key step and a major challenge in metagenomic research. Both supervised and unsupervised machine learning methods have been employed in binning. Genome binning belonging to unsupervised method clusters contigs into individual genome bins by machine learning methods without the assistance of any reference databases. So far a lot of genome binning tools have emerged. Evaluating these genome tools is of great significance to microbiological research. In this study, we evaluate 15 genome binning tools containing 12 original binning tools and 3 refining binning tools by comparing the performance of these tools on chicken gut metagenomic datasets and the first CAMI challenge datasets. RESULTS: For chicken gut metagenomic datasets, original genome binner MetaBat, Groopm2 and Autometa performed better than other original binner, and MetaWrap combined the binning results of them generated the most high-quality genome bins. For CAMI datasets, Groopm2 achieved the highest purity (> 0.9) with good completeness (> 0.8), and reconstructed the most high-quality genome bins among original genome binners. Compared with Groopm2, MetaBat2 had similar performance with higher completeness and lower purity. Genome refining binners DASTool predicated the most high-quality genome bins among all genomes binners. Most genome binner performed well for unique strains. Nonetheless, reconstructing common strains still is a substantial challenge for all genome binner. CONCLUSIONS: In conclusion, we tested a set of currently available, state-of-the-art metagenomics hybrid binning tools and provided a guide for selecting tools for metagenomic binning by comparing range of purity, completeness, adjusted rand index, and the number of high-quality reconstructed bins. Furthermore, available information for future binning strategy were concluded.

Effects of earplugs and eye masks combined with relaxing music on sleep, melatonin and cortisol levels in ICU patients: a randomized controlled trial.

INTRODUCTION: Intensive care unit (ICU) environmental factors such as noise and light have been cited as important causes of sleep deprivation in critically ill patients. Previous studies indicated that using earplugs and eye masks can improve REM sleep in healthy subjects in simulated ICU environment, and improve sleep quality in ICU patients. This study aimed to determine the effects of using earplugs and eye masks with relaxing background music on sleep, melatonin and cortisol levels in ICU patients. METHODS: Fifty patients who underwent a scheduled cardiac surgery and were expected to stay at least 2 nights in Cardiac Surgical ICU (CSICU) were included. They were randomized to sleep with or without earplugs and eye masks combined with 30-minute relaxing music during the postoperative nights in CSICU. Urine was analyzed for nocturnal melatonin and cortisol levels. Subjective sleep quality was evaluated using the Chinese version of Richards-Campbell Sleep Questionnaire (a visual analog scale, ranging 0-100). RESULTS: Data from 45 patients (20 in intervention group, 25 in control group) were analyzed. Significant differences were found between groups in depth of sleep, falling asleep, awakenings, falling asleep again after awakening and overall sleep quality (P < 0.05). Perceived sleep quality was better in the intervention group. No group differences were found in urinary melatonin levels and cortisol levels for the night before surgery, and the first and second nights post-surgery (P > 0.05). The urinary melatonin levels of the first and second postoperative nights were significantly lower than those of the night before surgery (P = 0.01). The opposite pattern was seen with urinary cortisol levels (P = 0.00). CONCLUSION: This combination of non-pharmacological interventions is useful for promoting sleep in ICU adult patients; however, any influence on nocturnal melatonin levels and cortisol level may have been masked by several factors such as the timing of surgery, medication use and individual differences. Larger scale studies would be needed to examine the potential influences of these factors on biological markers and intervention efficacy on sleep. TRIAL REGISTRATION: Chinese Clinical Trial Registry: ChiCTR-IOR-14005511 . Registered 21 November 2014.

[Effect and mechanisms of wenyang yiqi huoxue recipe on the apoptosis of photoreceptor cells in retinal degeneration slow mice].

OBJECTIVE: To observe the effect of Wenyang Yiqi Huoxue Recipe (WYHR) on the apoptosis of photoreceptor cells in retinal degeneration slow (RDS) mice, and to investigate its molecular mechanisms. METHODS: RDS mice were randomly divided into the model group and the Chinese medicine group,and C57BL/6J mice were selected as the normal control group. Each group consisted of 4 female mice and 2 male mice. Mice in the Chinese medicine group were administered with WYHR (10 mg/g) by gastrogavage since mating. Baby mice drunk WYHR decoction instead of drinking water once they were born. The offspring were administrated with low dose WYHR decoction by gastrogavage from the 7th postnatal day, and the dose was increased to that for adult mice from the 21st postnatal day. Physiological saline was administrated to mice in the model group and the normal control group by gastrogavage. At 18, 28 and 48 postnatal days, electroretinogram (ERG) was used to evaluate the retina functional variation, and the apoptotic rate of photoreceptor cells was determined by TUNEL staining. HE staining was performed. The number of photoreceptor cells of the outer nuclear layer was calculated. Furthermore, effect of WYHR on Rhodopsin and basic fibroblast growth factor (bFGF) expression was examined using immunochemical assay. RESULTS: Compared with the model group, a- and b-wave latency and amplitude, as well as the bFGF expression sharply increased in the Max-ERG of the Chinese medicine group (P < 0.05, P < 0.01) at 18 postnatal days. At 28 and 48 postnatal days, a- and b-wave latency and amplitude sharply increased, photoreceptor cell layer numbers of the outer nuclear layer obviously increased, the apoptosis rate of retinal photoreceptor cells obviously decreased, expressions of Rhodopsin and bFGF in the Chinese medicine group significantly increased (P < 0.05, P < 0.01). CONCLUSION: WYHR could effectively inhibit the apoptosis of photoreceptor cells in RDS mice, which might be attributed to up-regulating bFGF expression.

Effects of earplugs and eye masks on nocturnal sleep, melatonin and cortisol in a simulated intensive care unit environment.

INTRODUCTION: Environmental stimulus, especially noise and light, is thought to disrupt sleep in patients in the intensive care unit (ICU). This study aimed to determine the physiological and psychological effects of ICU noise and light, and of earplugs and eye masks, used in these conditions in healthy subjects. METHODS: Fourteen subjects underwent polysomnography under four conditions: adaptation, baseline, exposure to recorded ICU noise and light (NL), and NL plus use of earplugs and eye masks (NLEE). Urine was analyzed for melatonin and cortisol levels. Subjects rated their perceived sleep quality, anxiety levels and perception of environmental stimuli. RESULTS: Subjects had poorer perceived sleep quality, more light sleep, longer rapid eye movement (REM) latency, less REM sleep when exposed to simulated ICU noise and light (P < 0.05). Nocturnal melatonin (P = 0.007) and cortisol secretion levels (P = 0.004) differed significantly by condition but anxiety levels did not (P = 0.06). Use of earplugs and eye masks resulted in more REM time, shorter REM latency, less arousal (P < 0.05) and elevated melatonin levels (P = 0.002). CONCLUSIONS: Earplugs and eye masks promote sleep and hormone balance in healthy subjects exposed to simulated ICU noise and light, making their promotion in ICU patients reasonable.

[Study on acting mechanism of shenqi yiqi drop pill for intervening irido-microangiopathy in diabetic rats].

OBJECTIVE: Use laser confocal microscopy overspeed camera technique and fluorescence albumin labeling to study the acting mechanism of Qishen Yiqi Drop Pill (QYDP) for intervening irido-microangiopathy (IMAP) in diabetic rats. METHODS: Rat model of diabetes mellitus type 1 (DM1) was established by intraperitoneal injection of streptozocin (STZ). The model rats were randomly divided into three groups, the treatment group, the model group and the control group. At the same time a normal control group was set up. The treatment group was medicated with QYDP (prepared into liquid form), and the control group with Duobeisi liquor (1 g/kg per day) for 10 months. The dynamic state of iris microcirculation in rats was observed using laser confocal microscopy overspeed camera. RESULTS: Compared with the treatment group, blood flow in iris of model rats was slower significantly (P < 0.01); the fluorescence density and leakage area of inside and outside iris vessels, and the iris vascular diameter were significantly higher in the model group than those in the treatment group (P < 0.01). CONCLUSION: QYDP has definite effect in improving iris microcirculation, which can accelerate the blood flow, inhibit the abnormal expansion of vessels and improve the increased iris micro-vascular permeability.

[Treatment of choroidal neovascularization with Chinese medicine huangban bianxing recipe].

OBJECTIVE: To investigate the efficacy of Chinese medicine Huangban Bianxing Recipe (HBR) for treatment of choroidal neovascularization (CNV). METHODS: Eighty patients (97 eyes) suffered from CNV were assigned to two groups, the treated group treated by HBR orally taken one dose every day and the control group treated with ginkgo leaf tablet 1 tablet (19.2 mg) thrice a day, all for 3 months. The best corrected visual acuity (BCVA), fundus hemorrhagic area were measured; the neo-genetic vascular leakage area determined by fluorescent fundus angiography (FFA); and the out-layer high reflective band thickness (OHRBT) estimated by optical coherent tomography (OCT) were recorded before and after treatment. RESULTS: After treatment, in the treated group, BCVA was improved, with an effective rate of 67.12%, while in the control group, it was 33.33%, the former was better than the latter significantly (P < 0.05). The effective rate in reducing fundus hemorrhagic area in the two groups (85.11% vs 47.62%) and that in the CNV closure showed by FFA (67.50% vs 37.50%) were also superior in the treated group (P < 0.05). Besides, OCT showed that the OHRBT reduced significantly after treatment in the treated group (P < 0.01). CONCLUSION: HBR can improve the vision, reduce the CNV leakage in the macular region, and advance the absorption of fundus hemorrhage in CNV patients.

Integration of Multi-omics Data for Gene Regulatory Network Inference and Application to Breast Cancer.

Underlying a cancer phenotype is a specific gene regulatory network that represents the complex regulatory relationships between genes. However, it remains a challenge to find cancer-related gene regulatory network because of insufficient sample sizes and complex regulatory mechanisms in which gene is influenced by not only other genes but also other biological factors. With the development of high-throughput technologies and the unprecedented wealth of multi-omics data give us a new opportunity to design machine learning method to investigate underlying gene regulatory network. In this paper, we propose an approach, which use biweight midcorrelation to measure the correlation between factors and make use of nonconvex penalty based sparse regression for gene regulatory network inference (BMNPGRN). BMNCGRN incorporates multi-omics data (including DNA methylation and copy number variation) and their interactions in gene regulatory network model. The experimental results on synthetic datasets show that BMNPGRN outperforms popular and state-of-the-art methods (including DCGRN, ARACNE and CLR) under false positive control. Furthermore, we applied BMNPGRN on breast cancer (BRCA) data from The Cancer Genome Atlas database and provided gene regulatory network.