The rapid development of proteomics studies has resulted in large volumes of experimental data. The emergence of big data platform provides the opportunity to handle these large amounts of data. The integrated proteome resource, iProX (https://www.iprox.cn), which was initiated in 2017, has been greatly improved with an up-to-date big data platform implemented in 2021. Here, we describe the main iProX developments since its first publication in Nucleic Acids Research in 2019. First, a hyper-converged architecture with high scalability supports the submission process. A hadoop cluster can store large amounts of proteomics datasets, and a distributed, RESTful-styled Elastic Search engine can query millions of records within one second. Also, several new features, including the Universal Spectrum Identifier (USI) mechanism proposed by ProteomeXchange, RESTful Web Service API, and a high-efficiency reanalysis pipeline, have been added to iProX for better open data sharing. By the end of August 2021, 1526 datasets had been submitted to iProX, reaching a total data volume of 92.42TB. With the implementation of the big data platform, iProX can support PB-level data storage, hundreds of billions of spectra records, and second-level latency service capabilities that meet the requirements of the fast growing field of proteomics.

Constructing proper representations of molecules lies at the core of numerous tasks such as molecular property prediction and drug design. Graph neural networks, especially message passing neural network (MPNN) and its variants, have recently made remarkable achievements in molecular graph modeling. Albeit powerful, the one-sided focuses on atom (node) or bond (edge) information of existing MPNN methods lead to the insufficient representations of the attributed molecular graphs. Herein, we propose a Communicative Message Passing Neural Network (CMPNN) to improve the molecular embedding by strengthening the message interactions between nodes and edges through a communicative kernel. In addition, the message generation process is enriched by introducing a new message booster module. Extensive experiments demonstrated that the proposed model obtained superior performances against state-of-the-art baselines on six chemical property datasets. Further visualization also showed better representation capacity of our model.

Purpose The purpose of this study was to expedite the contouring process for MRI‐guided adaptive radiotherapy (MR‐IGART), a convolutional neural network (CNN) deep‐learning (DL) model is proposed to accurately segment the liver, kidneys, stomach, bowel and duodenum in 3D MR images. Methods Images and structure contours for 120 patients were collected retrospectively. Treatment sites included pancreas, liver, stomach, adrenal gland, and prostate. The proposed DL model contains a voxel‐wise label prediction CNN and a correction network which consists of two sub‐networks. The prediction CNN and sub‐networks in the correction network each includes a dense block which consists of twelve densely connected convolutional layers. The correction network was designed to improve the voxel‐wise labeling accuracy of a CNN by learning and enforcing implicit anatomical constraints in the segmentation process. Its sub‐networks learn to fix the erroneous classification of its previous network by taking as input both the original images and the softmax probability maps generated from its previous sub‐network. The parameters of each sub‐network were trained independently using piecewise training. The model was trained on 100 datasets, validated on 10 datasets and tested on the remaining 10 datasets. Dice coefficient, Hausdorff distance (HD) were calculated to evaluate the segmentation accuracy. Results The proposed DL model was able to segment the organs with good accuracy. The correction network outperformed the conditional random field (CRF), a most comparable method that is usually applied as a post‐processing step. For the 10 testing patients, the average Dice coefficients were 95.3 ± 0.73, 93.1 ± 2.22, 85.0 ± 3.75, 86.6 ± 2.69, and 65.5 ± 8.90 for liver, kidneys, stomach, bowel, and duodenum, respectively. The mean Hausdorff Distance (HD) were 5.41 ± 2.34, 6.23 ± 4.59, 6.88 ± 4.89, 5.90 ± 4.05, and 7.99 ± 6.84 mm, respectively. Manual contouring, as to correct the automatic segmentation results, was four times as fast as manual contouring from scratch. Conclusion The proposed method can automatically segment the liver, kidneys, stomach, bowel, and duodenum in 3D MR images with good accuracy. It is useful to expedite the manual contouring for MR‐IGART.

Purpose To extend the intravoxel incoherent motion (IVIM) magnetic resonance imaging (MRI) model to restricted diffusion and to simultaneously quantify the perfusion and restricted diffusion parameters in neck nodal metastases. Materials and Methods The non-Gaussian (NG)-IVIM model was developed and tested on diffusion-weighted MRI data collected on a 1.5-Tesla MRI scanner from 8 patients with head and neck cancer. Voxel-wise parameter quantification was performed by using a noise-rectified least-square fitting method. The NG-IVIM, IVIM, Kurtosis, and ADC (apparent diffusion coefficient) models were used for comparison. For each voxel, within the metastatic node, the optimal model was determined using the Bayesian Information Criterion. The voxel percentage preferred by each model was calculated and the optimal model map was generated. Monte Carlo simulations were performed to evaluate the accuracy and precision dependency of the new model. Results For the 8 neck nodes, the range of voxel percentage preferred by the NG-IVIM model was 2.3% - 79.3%. The optimal modal maps showed heterogeneities within the tumors. The Monte Carlo simulations demonstrated that the accuracy and precision of the NG-IVIM model improved by increasing signal-to-noise ratio and b value. Conclusion The NG-IVIM model characterizes perfusion and restricted diffusion simultaneously in neck nodal metastases.

No abstract