Online monitoring of air activation at the China spallation neutron source.

During the operation of high-energy proton accelerators, the air in the tunnel is activated with the production of radionuclides. For CSNS (China Spallation Neutron Source), the first pulsed neutron source in China for multidisciplinary research, an online air activation monitoring system was developed to evaluate the radiation safety of the staff and the public, which consisted of a NaI detector, Pb shielding, an MB container and a control system. With the monitoring system, gamma spectra of the activated air from controlled areas are measured, and the activity concentration and immersion dose rates of radionuclides in air are calculated and displayed in real time. The system has been in stable operation since February 2020, and results have been obtained for the evaluation of the radiation risk from activated air.

Radionuclides in target station coolant in the China Spallation Neutron Source.

The China Spallation Neutron Source (CSNS) is the first pulsed neutron source in China for multidisciplinary research. After operation with 80 kW proton beam for 4 months, 3 circuits of target station coolant, light water 1/2/3 were sampled, and radionuclides in coolants were measured. The results showed that, activity concentration of H-3 in coolant can be up to the magnitude of 1.00E+06 Bq/L, and the H-3 amount in light water 1 was the highest and the amount in light water 3 was the lowest, agreeing with the radiation field exposed by coolants. For Be-7, due to the complicated filtering and trapping process, amount of Be-7 in coolant differed from a minimum of 7.15E+01 Bq/L to a maximum of 4.58E+03 Bq/L. Comparison of the results with former measurements and simulated results were conducted. Permitted volumes for coolant discharge were presented. And the work time in the equipment room of cooling system after the beam is shut off is safe. Results in this research could provide reference data and measurement methods for similar accelerator devices.

RESEARCH ON DOSE MONITORING IN BACKSCATTERING NEUTRON HALL IN CHINA SPALLATION NEUTRON SOURCE.

The back-n project in China Spallation Neutron Source (CSNS) was launched primarily for nuclear data measurements. In the backscattering neutron hall, the neutron and gamma monitors were used for dose monitoring. Because of the dead time problem of monitors, performance of the monitors in such pulsed radiation field needs to be analyzed. In this research, experiments with dose monitors and personal dosemeters were conducted, and simulation by Monte Carlo code FLUKA was performed. Results showed that the values by monitors are smaller, and the larger the dose, the larger the difference. The reasons in term of energy response and dead time have been analyzed, and corrections were discussed. After corrections, the measured value can agree with the simulation results in the range of about a factor 3. Totally speaking, the values recorded by neutron and gamma monitors can be a reference for radiation safety management in CSNS.

Transcriptome analysis provides insights into the regulation of metabolic processes during postharvest cold storage of loquat (Eriobotrya japonica) fruit.

Loquat (Eriobotrya japonica) fruit accumulates lignin during postharvest storage under chilling conditions (0 °C), while low-temperature conditioning (LTC; 5 °C for 6 days followed by transfer to 0 °C) or heat treatment (HT; 40 °C for 4 h followed by transfer to 0 °C) can alleviate lignification. Here we compared transcriptome profiles of loquat fruit samples under LTC or HT to those stored at 0 °C at five time points from day 1 to day 8 after treatment. High-throughput transcriptome sequences were de novo assembled into 53,319 unique transcripts with an N50 length of 1306 bp. A total of 2235 differentially expressed genes were identified in LTC, and 1020 were identified in HT compared to 0 °C. Key genes in the lignin biosynthetic pathway, including EjPAL2, EjCAD1, EjCAD3, 4CL, COMT, and HCT, were responsive to LTC or HT treatment, but they showed different expression patterns during the treatments, indicating that different structural genes could regulate lignification at different treatment stages. Coexpression network analysis showed that these candidate biosynthetic genes were associated with a number of transcription factors, including those belonging to the AP2, MYB, and NAC families. Gene ontology (GO) enrichment analysis of differentially expressed genes indicated that biological processes such as stress responses, cell wall and lignin metabolism, hormone metabolism, and metal ion transport were significantly affected under LTC or HT treatment when compared to 0 °C. Our analyses provide insights into transcriptome responses to postharvest treatments in loquat fruit.

DriverML: a machine learning algorithm for identifying driver genes in cancer sequencing studies.

Although rapid progress has been made in computational approaches for prioritizing cancer driver genes, research is far from achieving the ultimate goal of discovering a complete catalog of genes truly associated with cancer. Driver gene lists predicted from these computational tools lack consistency and are prone to false positives. Here, we developed an approach (DriverML) integrating Rao's score test and supervised machine learning to identify cancer driver genes. The weight parameters in the score statistics quantified the functional impacts of mutations on the protein. To obtain optimized weight parameters, the score statistics of prior driver genes were maximized on pan-cancer training data. We conducted rigorous and unbiased benchmark analysis and comparisons of DriverML with 20 other existing tools in 31 independent datasets from The Cancer Genome Atlas (TCGA). Our comprehensive evaluations demonstrated that DriverML was robust and powerful among various datasets and outperformed the other tools with a better balance of precision and sensitivity. In vitro cell-based assays further proved the validity of the DriverML prediction of novel driver genes. In summary, DriverML uses an innovative, machine learning-based approach to prioritize cancer driver genes and provides dramatic improvements over currently existing methods. Its source code is available at https://github.com/HelloYiHan/DriverML.

A comprehensive evaluation of alignment software for reduced representation bisulfite sequencing data.

Motivation: The rapid development of next-generation sequencing technology provides an opportunity to study genome-wide DNA methylation at single-base resolution. However, depletion of unmethylated cytosines brings challenges for aligning bisulfite-converted sequencing reads to a large reference. Software tools for aligning methylation reads have not yet been comprehensively evaluated, especially for the widely used reduced representation bisulfite sequencing (RRBS) that involves enrichment for CpG islands (CGIs). Results: We specially developed a simulator, RRBSsim, for benchmarking analysis of RRBS data. We performed extensive comparison of seven mapping algorithms for methylation analysis in both real and simulated RRBS data. Eighteen lung tumors and matched adjacent tissues were sequenced by the RRBS protocols. Our empirical evaluation found that methylation results were less consistent between software tools for CpG sites with low sequencing depth, medium methylation level, on CGI shores or gene body. These observations were further confirmed by simulations that indicated software tools generally had lower recall of detecting these vulnerable CpG sites and lower precision of estimating methylation levels in these CpG sites. Among the software tools tested, bwa-meth and BS-Seeker2 (bowtie2) are currently our preferred aligners for RRBS data in terms of recall, precision and speed. Existing aligners cannot efficiently handle moderately methylated CpG sites and those CpG sites on CGI shores or gene body. Interpretation of methylation results from these vulnerable CpG sites should be treated with caution. Our study reveals several important features inherent in methylation data, and RRBSsim provides guidance to advance sequence-based methylation data analysis and methodological development. Availability and implementation: RRBSsim is a simulator for benchmarking analysis of RRBS data and its source code is available at https://github.com/xwBio/RRBSsim or https://github.com/xwBio/Docker-RRBSsim. Supplementary information: Supplementary data are available at Bioinformatics online.

A Reweighted Scheme to Improve the Representation of the Neural Autoregressive Distribution Estimator.

The neural autoregressive distribution estimator(NADE) is a competitive model for the task of density estimation in the field of machine learning. While NADE mainly focuses on the problem of estimating density, the ability for dealing with other tasks remains to be improved. In this paper, we introduce a simple and efficient reweighted scheme to modify the parameters of the learned NADE. We make use of the structure of NADE, and the weights are derived from the activations in the corresponding hidden layers. The experiments show that the features from unsupervised learning with our reweighted scheme would be more meaningful, and the performance of the initialization for neural networks has a significant improvement as well.

Shape Completion Using Deep Boltzmann Machine.

Shape completion is an important task in the field of image processing. An alternative method is to capture the shape information and finish the completion by a generative model, such as Deep Boltzmann Machine. With its powerful ability to deal with the distribution of the shapes, it is quite easy to acquire the result by sampling from the model. In this paper, we make use of the hidden activation of the DBM and incorporate it with the convolutional shape features to fit a regression model. We compare the output of the regression model with the incomplete shape feature in order to set a proper and compact mask for sampling from the DBM. The experiment shows that our method can obtain realistic results without any prior information about the incomplete object shape.

Study on patient-induced radioactivity during proton treatment in hengjian proton medical facility.

At present, increasingly more proton medical facilities have been established globally for better curative effect and less side effect in tumor treatment. Compared with electron and photon, proton delivers more energy and dose at its end of range (Bragg peak), and has less lateral scattering for its much larger mass. However, proton is much easier to produce neutron and induced radioactivity, which makes radiation protection for proton accelerators more difficult than for electron accelerators. This study focuses on the problem of patient-induced radioactivity during proton treatment, which has been ignored for years. However, we confirmed it is a vital factor for radiation protection to both patient escort and positioning technician, by FLUKA's simulation and activation formula calculation of Hengjian Proton Medical Facility (HJPMF), whose energy ranges from 130 to 230MeV. Furthermore, new formulas for calculating the activity buildup process of periodic irradiation were derived and used to study the relationship between saturation degree and half-life of nuclides. Finally, suggestions are put forward to lessen the radiation hazard from patient-induced radioactivity.

Semilocal convergence theorem for the inverse-free Jarratt method under new Hölder conditions.

Under the new Hölder conditions, we consider the convergence analysis of the inverse-free Jarratt method in Banach space which is used to solve the nonlinear operator equation. We establish a new semilocal convergence theorem for the inverse-free Jarratt method and present an error estimate. Finally, three examples are provided to show the application of the theorem.