Development of a regional-based predictive model of incidence of traumatic spinal cord injury using machine learning algorithms.

Objective: To develop a predictive model of incidence of traumatic spinal cord injury (TSCI). Methods: The data for training the model included both the incidence data and the covariates. The incidence data were extracted from systematic reviews and the covariates were extracted from data available in the international road federation database. Then the feature processing measures were taken. First we defined a hyper-parameter, missing-value threshold, in order to eliminate features that exceed this threshold. To tackle the problem of overfitting of model we determined the Pearson correlation of features and excluded those with more than 0.7 correlation. After feature selection three different models including simple linear regression, support vector regression, and multi-layer perceptron were examined to fit the purposes of this study. Finally, we evaluated the model based on three standard metrics: Mean Absolute Error, Root Mean Square Error, and R2. Results: Our machine-learning based model could predict the incidence rate of TSCI with the mean absolute error of 4.66. Our model found "Vehicles in use, Total vehicles/Km of roads", "Injury accidents/100 Million Veh-Km", "Vehicles in use, Vans, Pick-ups, Lorries, Road Tractors", "Inland surface Passengers Transport (Mio Passenger-Km), Rail", and "% paved" as top predictors of transport-related TSCI (TRTSCI). Conclusions: Our model is proved to have a high accuracy to predict the incidence rate of TSCI for countries, especially where the main etiology of TSCI is related to road traffic injuries. Using this model, we can help the policymakers for resource allocation and evaluation of preventive measures.

TumorDecon: A digital cytometry software.

There are many experimental methods for characterizing immune profiles of tumors, such as flow and mass cytometry. However, these approaches are time and resource intensive. Thus, several "digital cytometry" methods have been developed to extract cell frequencies from RNA-seq data. Here, we introduce TumorDecon, named for its potential to deconvolve the distribution of cells from the gene expression levels of a bulk of cells, such as a tumor. The Python package provides an accessible way of applying these methods. It includes four deconvolution methods as well as several gene sets, signature matrices, and functions for generating custom signature matrices.

SVNN: an efficient PacBio-specific pipeline for structural variations calling using neural networks.

BACKGROUND: Once aligned, long-reads can be a useful source of information to identify the type and position of structural variations. However, due to the high sequencing error of long reads, long-read structural variation detection methods are far from precise in low-coverage cases. To be accurate, they need to use high-coverage data, which in turn, results in an extremely time-consuming pipeline, especially in the alignment phase. Therefore, it is of utmost importance to have a structural variation calling pipeline which is both fast and precise for low-coverage data. RESULTS: In this paper, we present SVNN, a fast yet accurate, structural variation calling pipeline for PacBio long-reads that takes raw reads as the input and detects structural variants of size larger than 50 bp. Our pipeline utilizes state-of-the-art long-read aligners, namely NGMLR and Minimap2, and structural variation callers, videlicet Sniffle and SVIM. We found that by using a neural network, we can extract features from Minimap2 output to detect a subset of reads that provide useful information for structural variation detection. By only mapping this subset with NGMLR, which is far slower than Minimap2 but better serves downstream structural variation detection, we can increase the sensitivity in an efficient way. As a result of using multiple tools intelligently, SVNN achieves up to 20 percentage points of sensitivity improvement in comparison with state-of-the-art methods and is three times faster than a naive combination of state-of-the-art tools to achieve almost the same accuracy. CONCLUSION: Since prohibitive costs of using high-coverage data have impeded long-read applications, with SVNN, we provide the users with a much faster structural variation detection platform for PacBio reads with high precision and sensitivity in low-coverage scenarios.

Immune classification of clear cell renal cell carcinoma.

Since the outcome of treatments, particularly immunotherapeutic interventions, depends on the tumor immune micro-environment (TIM), several experimental and computational tools such as flow cytometry, immunohistochemistry, and digital cytometry have been developed and utilized to classify TIM variations. In this project, we identify immune pattern of clear cell renal cell carcinomas (ccRCC) by estimating the percentage of each immune cell type in 526 renal tumors using the new powerful technique of digital cytometry. The results, which are in agreement with the results of a large-scale mass cytometry analysis, show that the most frequent immune cell types in ccRCC tumors are CD8+ T-cells, macrophages, and CD4+ T-cells. Saliently, unsupervised clustering of ccRCC primary tumors based on their relative number of immune cells indicates the existence of four distinct groups of ccRCC tumors. Tumors in the first group consist of approximately the same numbers of macrophages and CD8+ T-cells and and a slightly smaller number of CD4+ T cells than CD8+ T cells, while tumors in the second group have a significantly high number of macrophages compared to any other immune cell type (P-value [Formula: see text]). The third group of ccRCC tumors have a significantly higher number of CD8+ T-cells than any other immune cell type (P-value [Formula: see text]), while tumors in the group 4 have approximately the same numbers of macrophages and CD4+ T-cells and a significantly smaller number of CD8+ T-cells than CD4+ T-cells (P-value [Formula: see text]). Moreover, there is a high positive correlation between the expression levels of IFNG and PDCD1 and the percentage of CD8+ T-cells, and higher stage and grade of tumors have a substantially higher percentage of CD8+ T-cells. Furthermore, the primary tumors of patients, who are tumor free at the last time of follow up, have a significantly higher percentage of mast cells (P-value [Formula: see text]) compared to the patients with tumors for all groups of tumors except group 3.

Data Driven Mathematical Model of Colon Cancer Progression.

Every colon cancer has its own unique characteristics, and therefore may respond differently to identical treatments. Here, we develop a data driven mathematical model for the interaction network of key components of immune microenvironment in colon cancer. We estimate the relative abundance of each immune cell from gene expression profiles of tumors, and group patients based on their immune patterns. Then we compare the tumor sensitivity and progression in each of these groups of patients, and observe differences in the patterns of tumor growth between the groups. For instance, in tumors with a smaller density of naive macrophages than activated macrophages, a higher activation rate of macrophages leads to an increase in cancer cell density, demonstrating a negative effect of macrophages. Other tumors however, exhibit an opposite trend, showing a positive effect of macrophages in controlling tumor size. Although the results indicate that for all patients the size of the tumor is sensitive to the parameters related to macrophages, such as their activation and death rate, this research demonstrates that no single biomarker could predict the dynamics of tumors.