Cancer prognosis with shallow tumor RNA sequencing

Advances in long-read single-cell transcriptomics

Long-read single-cell transcriptomics (scRNA-Seq) is revolutionizing the way we profile heterogeneity in disease. Traditional short-read scRNA-Seq methods are limited in their ability to provide complete transcript coverage, resolve isoforms, and identify novel transcripts. The scRNA-Seq protocols developed for long-read sequencing platforms overcome these limitations by enabling the characterization of full-length transcripts. Long-read scRNA-Seq techniques initially suffered from comparatively poor accuracy compared to short read scRNA-Seq. However, with improvements in accuracy, accessibility, and cost efficiency, long-reads are gaining popularity in the field of scRNA-Seq. This review details the advances in long-read scRNA-Seq, with an emphasis on library preparation protocols and downstream bioinformatics analysis tools.

Identification of BGN positive fibroblasts as a driving factor for colorectal cancer and development of its related prognostic model combined with machine learning

BACKGROUND

Numerous studies have indicated that cancer-associated fibroblasts (CAFs) play a crucial role in the progression of colorectal cancer (CRC). However, there are still many unknowns regarding the exact role of CAF subtypes in CRC.

METHODS

The data for this study were obtained from bulk, single-cell, and spatial transcriptomic sequencing data. Bioinformatics analysis, in vitro experiments, and machine learning methods were employed to investigate the functional characteristics of CAF subtypes and construct prognostic models.

RESULTS

Our study demonstrates that Biglycan (BGN) positive cancer-associated fibroblasts (BGN + Fib) serve as a driver in colorectal cancer (CRC). The proportion of BGN + Fib increases gradually with the progression of CRC, and high infiltration of BGN + Fib is associated with poor prognosis in terms of overall survival (OS) and recurrence-free survival (RFS) in CRC. Downregulation of BGN expression in cancer-associated fibroblasts (CAFs) significantly reduces migration and proliferation of CRC cells. Among 101 combinations of 10 machine learning algorithms, the StepCox[both] + plsRcox combination was utilized to develop a BGN + Fib derived risk signature (BGNFRS). BGNFRS was identified as an independent adverse prognostic factor for CRC OS and RFS, outperforming 92 previously published risk signatures. A Nomogram model constructed based on BGNFRS and clinical-pathological features proved to be a valuable tool for predicting CRC prognosis.

CONCLUSION

In summary, our study identified BGN + Fib as drivers of CRC, and the derived BGNFRS was effective in predicting the OS and RFS of CRC patients.

Stratification of Colorectal Patients Based on Survival Analysis Shows the Value of Consensus Molecular Subtypes and Reveals the CBLL1 Gene as a Biomarker of CMS2 Tumours

The consensus molecular subtypes (CMSs) classification of colorectal cancer (CRC) is a system for patient stratification that can be potentially applied to therapeutic decisions. Hakai (CBLL1) is an E3 ubiquitin-ligase that induces the ubiquitination and degradation of E-cadherin, inducing epithelial-to-mesenchymal transition (EMT), tumour progression and metastasis. Using bioinformatic methods, we have analysed CBLL1 expression on a large integrated cohort of primary tumour samples from CRC patients. The cohort included survival data and was divided into consensus molecular subtypes. Colon cancer tumourspheres were used to analyse the expression of stem cancer cells markers via RT-PCR and Western blotting. We show that CBLL1 gene expression is specifically associated with canonical subtype CMS2. WNT target genes LGR5 and c-MYC show a similar association with CMS2 as CBLL1. These mRNA levels are highly upregulated in cancer tumourspheres, while CBLL1 silencing shows a clear reduction in tumoursphere size and in stem cell biomarkers. Importantly, CMS2 patients with high CBLL1 expression displayed worse overall survival (OS), which is similar to that associated with CMS4 tumours. Our findings reveal CBLL1 as a specific biomarker for CMS2 and the potential of using CMS2 with high CBLL1 expression to stratify patients with poor OS.

Clinically-Inspired Multi-Agent Transformers for Disease Trajectory Forecasting From Multimodal Data

Deep neural networks are often applied to medical images to automate the problem of medical diagnosis. However, a more clinically relevant question that practitioners usually face is how to predict the future trajectory of a disease. Current methods for prognosis or disease trajectory forecasting often require domain knowledge and are complicated to apply. In this paper, we formulate the prognosis prediction problem as a one-to-many prediction problem. Inspired by a clinical decision-making process with two agents-a radiologist and a general practitioner - we predict prognosis with two transformer-based components that share information with each other. The first transformer in this framework aims to analyze the imaging data, and the second one leverages its internal states as inputs, also fusing them with auxiliary clinical data. The temporal nature of the problem is modeled within the transformer states, allowing us to treat the forecasting problem as a multi-task classification, for which we propose a novel loss. We show the effectiveness of our approach in predicting the development of structural knee osteoarthritis changes and forecasting Alzheimer's disease clinical status directly from raw multi-modal data. The proposed method outperforms multiple state-of-the-art baselines with respect to performance and calibration, both of which are needed for real-world applications. An open-source implementation of our method is made publicly available at https://github.com/Oulu-IMEDS/CLIMATv2.

SaBrcada: Survival Intervals Prediction for Breast Cancer Patients by Dimension Raising and Age Stratification

(1) Background: Breast cancer is the second leading cause of cancer death among women. The accurate prediction of survival intervals will help physicians make informed decisions about treatment strategies or the use of palliative care. (2) Methods: Gene expression is predictive and correlates to patient prognosis. To establish a reliable prediction tool, we collected a total of 1187 RNA-seq data points from breast cancer patients (median age 58 years) in Fragments Per Kilobase Million (FPKM) format from the TCGA database. Among them, we selected 144 patients with date of death information to establish the SaBrcada-AD dataset. We first normalized the SaBrcada-AD dataset to TPM to build the survival prediction model SaBrcada. After normalization and dimension raising, we used the differential gene expression data to test eight different deep learning architectures. Considering the effect of age on prognosis, we also performed a stratified random sampling test on all ages between the lower and upper quartiles of patient age, 48 and 69 years; (3) Results: Stratifying by age 61, the performance of SaBrcada built by GoogLeNet was improved to a highest accuracy of 0.798. We also built a free website tool to provide five predicted survival periods: within six months, six months to one year, one to three years, three to five years, or over five years, for clinician reference. (4) Conclusions: We built the prediction model, SaBrcada, and the website tool of the same name for breast cancer survival analysis. Through these models and tools, clinicians will be provided with survival interval information as a basis for formulating precision medicine.

Systematic assessment of prognostic molecular features across cancers

Precision oncology promises accurate prediction of disease trajectories by utilizing molecular features of tumors. We present a systematic analysis of the prognostic potential of diverse molecular features across large cancer cohorts. We find that the mRNA expression of biologically coherent sets of genes (modules) is substantially more predictive of patient survival than single-locus genomic and transcriptomic aberrations. Extending our analysis beyond existing curated gene modules, we find a large novel class of highly prognostic DNA/RNA cis-regulatory modules associated with dynamic gene expression within cancers. Remarkably, in more than 82% of cancers, modules substantially improve survival stratification compared with conventional clinical factors and prominent genomic aberrations. The prognostic potential of cancer modules generalizes to external cohorts better than conventionally used single-gene features. Finally, a machine-learning framework demonstrates the combined predictive power of multiple modules, yielding prognostic models that perform substantially better than existing histopathological and clinical factors in common use.

Transcriptome profiling for precision cancer medicine using shallow nanopore cDNA sequencing

Transcriptome profiling is a mainstay of translational cancer research and is increasingly finding its way into precision oncology. While bulk RNA sequencing (RNA-seq) is widely available, high investment costs and long data return time are limiting factors for clinical applications. We investigated a portable nanopore long-read sequencing device (MinION, Oxford Nanopore Technologies) for transcriptome profiling of tumors. In particular, we investigated the impact of lower coverage than that of larger sequencing devices by comparing shallow nanopore RNA-seq data with short-read RNA-seq data generated using reversible dye terminator technology (Illumina) for ten samples representing four cancer types. Coupled with ShaNTi (Shallow Nanopore sequencing for Transcriptomics), a newly developed data processing pipeline, a turnaround time of five days was achieved. The correlation of normalized gene-level counts between nanopore and Illumina RNA-seq was high for MinION but not for very low-throughput Flongle flow cells (r = 0.89 and r = 0.24, respectively). A cost-saving approach based on multiplexing of four samples per MinION flow cell maintained a high correlation with Illumina data (r = 0.56-0.86). In addition, we compared the utility of nanopore and Illumina RNA-seq data for analysis tools commonly applied in translational oncology: (1) Shallow nanopore and Illumina RNA-seq were equally useful for inferring signaling pathway activities with PROGENy. (2) Highly expressed genes encoding kinases targeted by clinically approved small-molecule inhibitors were reliably identified by shallow nanopore RNA-seq. (3) In tumor microenvironment composition analysis, quanTIseq performed better than CIBERSORT, likely due to higher average expression of the gene set used for deconvolution. (4) Shallow nanopore RNA-seq was successfully applied to detect fusion genes using the JAFFAL pipeline. These findings suggest that shallow nanopore RNA-seq enables rapid and biologically meaningful transcriptome profiling of tumors, and warrants further exploration in precision cancer medicine studies.

Optimal microRNA Sequencing Depth to Predict Cancer Patient Survival with Random Forest and Cox Models

(1) Background: tumor profiling enables patient survival prediction. The two essential parameters to be calibrated when designing a study based on tumor profiles from a cohort are the sequencing depth of RNA-seq technology and the number of patients. This calibration is carried out under cost constraints, and a compromise has to be found. In the context of survival data, the goal of this work is to benchmark the impact of the number of patients and of the sequencing depth of miRNA-seq and mRNA-seq on the predictive capabilities for both the Cox model with elastic net penalty and random survival forest. (2) Results: we first show that the Cox model and random survival forest provide comparable prediction capabilities, with significant differences for some cancers. Second, we demonstrate that miRNA and/or mRNA data improve prediction over clinical data alone. mRNA-seq data leads to slightly better prediction than miRNA-seq, with the notable exception of lung adenocarcinoma for which the tumor miRNA profile shows higher predictive power. Third, we demonstrate that the sequencing depth of RNA-seq data can be reduced for most of the investigated cancers without degrading the prediction abilities, allowing the creation of independent validation sets at a lower cost. Finally, we show that the number of patients in the training dataset can be reduced for the Cox model and random survival forest, allowing the use of different models on different patient subgroups.

A novel prognostic model for cutaneous melanoma based on an immune-related gene signature and clinical variables

Abundant evidence has indicated that the prognosis of cutaneous melanoma (CM) patients is highly complicated by the tumour immune microenvironment. We retrieved the clinical data and gene expression data of CM patients in The Cancer Genome Atlas (TCGA) database for modelling and validation analysis. Based on single-sample gene set enrichment analysis (ssGSEA) and consensus clustering analysis, CM patients were classified into three immune level groups, and the differences in the tumour immune microenvironment and clinical characteristics were evaluated. Seven immune-related CM prognostic molecules, including three mRNAs (SUCO, BTN3A1 and TBC1D2), three lncRNAs (HLA-DQB1-AS1, C9orf139 and C22orf34) and one miRNA (hsa-miR-17-5p), were screened by differential expression analysis, ceRNA network analysis, LASSO Cox regression analysis and univariate Cox regression analysis. Their biological functions were mainly concentrated in the phospholipid metabolic process, transcription regulator complex, protein serine/threonine kinase activity and MAPK signalling pathway. We established a novel prognostic model for CM integrating clinical variables and immune molecules that showed promising predictive performance demonstrated by receiver operating characteristic curves (AUC ≥ 0.74), providing a scientific basis for predicting the prognosis and improving the clinical outcomes of CM patients.

High-temporal resolution profiling reveals distinct immune trajectories following the first and second doses of COVID-19 mRNA vaccines

Knowledge of the mechanisms underpinning the development of protective immunity conferred by mRNA vaccines is fragmentary. Here, we investigated responses to coronavirus disease 2019 (COVID-19) mRNA vaccination via high-temporal resolution blood transcriptome profiling. The first vaccine dose elicited modest interferon and adaptive immune responses, which peaked on days 2 and 5, respectively. The second vaccine dose, in contrast, elicited sharp day 1 interferon, inflammation, and erythroid cell responses, followed by a day 5 plasmablast response. Both post-first and post-second dose interferon signatures were associated with the subsequent development of antibody responses. Yet, we observed distinct interferon response patterns after each of the doses that may reflect quantitative or qualitative differences in interferon induction. Distinct interferon response phenotypes were also observed in patients with COVID-19 and were associated with severity and differences in duration of intensive care. Together, this study also highlights the benefits of adopting high-frequency sampling protocols in profiling vaccine-elicited immune responses.

SurviveAI: Long Term Survival Prediction of Cancer Patients Based on Somatic RNA-Seq Expression

Motivation

Prediction of cancer outcome is a major challenge in oncology and is essential for treatment planning. Repositories such as The Cancer Genome Atlas (TCGA) contain vast amounts of data for many types of cancers. Our goal was to create reliable prediction models using TCGA data and validate them using an external dataset.

Results

For 16 TCGA cancer type cohorts we have optimized a Random Forest prediction model using parameter grid search followed by a backward feature elimination loop for dimensions reduction. For each feature that was removed, the model was retrained and the area under the curve of the receiver operating characteristic (AUC-ROC) was calculated using test data. Five prediction models gave AUC-ROC bigger than 80%. We used Clinical Proteomic Tumor Analysis Consortium v3 (CPTAC3) data for validation. The most enriched pathways for the top models were those involved in basic functions related to tumorigenesis and organ development. Enrichment for 2 prediction models of the TCGA-KIRP cohort was explored, one with 42 genes (AUC-ROC = 0.86) the other is composed of 300 genes (AUC-ROC = 0.85). The most enriched networks for both models share only 5 network nodes: DMBT1, IL11, HOXB6, TRIB3, PIM1. These genes play a significant role in renal cancer and might be used for prognosis prediction and as candidate therapeutic targets.

Availability And Implementation

The prediction models were created and tested using Python SciKit-Learn package. They are freely accessible via a friendly web interface we called surviveAI at https://tinyurl.com/surviveai.

Prognosis of lasso-like penalized Cox models with tumor profiling improves prediction over clinical data alone and benefits from bi-dimensional pre-screening

BACKGROUND

Prediction of patient survival from tumor molecular '-omics' data is a key step toward personalized medicine. Cox models performed on RNA profiling datasets are popular for clinical outcome predictions. But these models are applied in the context of "high dimension", as the number p of covariates (gene expressions) greatly exceeds the number n of patients and e of events. Thus, pre-screening together with penalization methods are widely used for dimensional reduction.

METHODS

In the present paper, (i) we benchmark the performance of the lasso penalization and three variants (i.e., ridge, elastic net, adaptive elastic net) on 16 cancers from TCGA after pre-screening, (ii) we propose a bi-dimensional pre-screening procedure based on both gene variability and p-values from single variable Cox models to predict survival, and (iii) we compare our results with iterative sure independence screening (ISIS).

RESULTS

First, we show that integration of mRNA-seq data with clinical data improves predictions over clinical data alone. Second, our bi-dimensional pre-screening procedure can only improve, in moderation, the C-index and/or the integrated Brier score, while excluding irrelevant genes for prediction. We demonstrate that the different penalization methods reached comparable prediction performances, with slight differences among datasets. Finally, we provide advice in the case of multi-omics data integration.

CONCLUSIONS

Tumor profiles convey more prognostic information than clinical variables such as stage for many cancer subtypes. Lasso and Ridge penalizations perform similarly than Elastic Net penalizations for Cox models in high-dimension. Pre-screening of the top 200 genes in term of single variable Cox model p-values is a practical way to reduce dimension, which may be particularly useful when integrating multi-omics.

Miller H, Hu X, Kumar R, T. Ngo P, Yan J, B. Frieboes H. Prediction of lung cancer immunotherapy response via machine learning analysis of immune cell lineage and surface markers

BACKGROUND: Although advances have been made in cancer immunotherapy, patient benefits remain elusive. For non-small cell lung cancer (NSCLC), monoclonal antibodies targeting programmed death-1 (PD-1) and programmed death ligand-1 (PD-L1) have shown survival benefit compared to chemotherapy. Personalization of treatment would be facilitated by a priori identification of patients likely to benefit. OBJECTIVE: This pilot study applied a suite of machine learning methods to analyze mass cytometry data of immune cell lineage and surface markers from blood samples of a small cohort (n= 13) treated with Pembrolizumab, Atezolizumab, Durvalumab, or Nivolumab as monotherapy. METHODS: Four different comparisons were evaluated between data collected at an initial visit (baseline), after 12-weeks of immunotherapy, and from healthy (control) samples: healthy vs patients at baseline, Responders vs Non-Responders at baseline, Healthy vs 12-week Responders, and Responders vs Non-Responders at 12-weeks. The algorithms Random Forest, Partial Least Squares Discriminant Analysis, Multi-Layer Perceptron, and Elastic Net were applied to find features differentiating between these groups and provide for the capability to predict outcomes. RESULTS: Particular combinations and proportions of immune cell lineage and surface markers were sufficient to accurately discriminate between the groups without overfitting the data. In particular, markers associated with the B-cell phenotype were identified as key features. CONCLUSIONS: This study illustrates a comprehensive machine learning analysis of circulating immune cell characteristics of NSCLC patients with the potential to predict response to immunotherapy. Upon further evaluation in a larger cohort, the proposed methodology could help guide personalized treatment selection in clinical practice.

Review of Multiple Myeloma Genetics including Effects on Prognosis, Response to Treatment, and Diagnostic Workup

Multiple myeloma is a disorder of the monoclonal plasma cells and is the second most common hematologic malignancy. Despite improvements in survival with newer treatment regimens, multiple myeloma remains an incurable disease and most patients experience multiple relapses. Multiple myeloma disease initiation and progression are highly dependent on complex genetic aberrations. This review will summarize the current knowledge of these genetic aberrations, how they affect prognosis and the response to treatment, and review sensitive molecular techniques for multiple myeloma workup, with the ultimate goal of detecting myeloma progression early, allowing for timely treatment initiation.

Predicting cancer prognosis and drug response from the tumor microbiome

Tumor gene expression is predictive of patient prognosis in some cancers. However, RNA-seq and whole genome sequencing data contain not only reads from host tumor and normal tissue, but also reads from the tumor microbiome, which can be used to infer the microbial abundances in each tumor. Here, we show that tumor microbial abundances, alone or in combination with tumor gene expression, can predict cancer prognosis and drug response to some extent-microbial abundances are significantly less predictive of prognosis than gene expression, although similarly as predictive of drug response, but in mostly different cancer-drug combinations. Thus, it appears possible to leverage existing sequencing technology, or develop new protocols, to obtain more non-redundant information about prognosis and drug response from RNA-seq and whole genome sequencing experiments than could be obtained from tumor gene expression or genomic data alone.

Assessment of the Risk of Severe Dengue Using Intrahost Viral Population in Dengue Virus Serotype 2 Patients via Machine Learning

Dengue virus, a positive-sense single-stranded RNA virus, continuously threatens human health. Although several criteria for evaluation of severe dengue have been recently established, the ability to prognose the risk of severe outcomes for dengue patients remains limited. Mutant spectra of RNA viruses, including single nucleotide variants (SNVs) and defective virus genomes (DVGs), contribute to viral virulence and growth. Here, we determine the potency of intrahost viral population in dengue patients with primary infection that progresses into severe dengue. A total of 65 dengue virus serotype 2 infected patients in primary infection including 17 severe cases were enrolled. We utilized deep sequencing to directly define the frequency of SNVs and detection times of DVGs in sera of dengue patients and analyzed their associations with severe dengue. Among the detected SNVs and DVGs, the frequencies of 9 SNVs and the detection time of 1 DVG exhibited statistically significant differences between patients with dengue fever and those with severe dengue. By utilizing the detected frequencies/times of the selected SNVs/DVG as features, the machine learning model showed high average with a value of area under the receiver operating characteristic curve (AUROC, 0.966 ± 0.064). The elevation of the frequency of SNVs at E (nucleotide position 995 and 2216), NS2A (nucleotide position 4105), NS3 (nucleotide position 4536, 4606), and NS5 protein (nucleotide position 7643 and 10067) and the detection times of the selected DVG that had a deletion junction in the E protein region (nucleotide positions of the junction: between 969 and 1022) increased the possibility of dengue patients for severe dengue. In summary, we demonstrated the detected frequencies/times of SNVs/DVG in dengue patients associated with severe disease and successfully utilized them to discriminate severe patients using machine learning algorithm. The identified SNVs and DVGs that are associated with severe dengue will expand our understanding of intrahost viral population in dengue pathogenesis.

Development and validation of a transcriptomic signature-based model as the predictive, preventive, and personalized medical strategy for preterm birth within 7 days in threatened preterm labor women

Preterm birth (PTB) is the leading cause of neonatal death. The essential strategy to prevent PTB is the accurate identification of threatened preterm labor (TPTL) women who will have PTB in a short time (< 7 days). Here, we aim to propose a clinical model to contribute to the effective prediction, precise prevention, and personalized medical treatment for PTB < 7 days in TPTL women through bioinformatics analysis and prospective cohort studies. In this study, the 1090 key genes involved in PTB < 7 days in the peripheral blood of TPTL women were ascertained using WGCNA. Based on this, the biological basis of immune-inflammatory activation (e.g., IFNγ and TNFα signaling) as well as immune cell disorders (e.g., monocytes and Th17 cells) in PTB < 7 days were revealed. Then, four core genes (JOSD1, IDNK, ZMYM3, and IL1B) that best represent their transcriptomic characteristics were screened by SVM and LASSO algorithm. Therefore, a prediction model with an AUC of 0.907 was constructed, which was validated in a larger population (AUC = 0.783). Moreover, the predictive value (AUC = 0.957) and clinical feasibility of this model were verified through the clinical prospective cohort we established. In conclusion, in the context of Predictive, Preventive, and Personalized Medicine (3PM), we have developed and validated a model to predict PTB < 7 days in TPTL women. This is promising to greatly improve the accuracy of clinical prediction, which would facilitate the personalized management of TPTL women to precisely prevent PTB < 7 days and improve maternal-fetal outcomes.

Construction and Comprehensive Prognostic Analysis of a Novel Immune-Related lncRNA Signature and Immune Landscape in Gastric Cancer

Gastric cancer (GC) is a malignant tumor with high mortality and poor prognosis. Immunotherapies, especially immune checkpoint inhibitors (ICI), are widely used in various tumors, but patients with GC do not benefit much from immunotherapies. Therefore, effective predictive biomarkers are urgently needed for GC patients to realize the benefits of immunotherapy. Recent studies have indicated that long noncoding RNAs (lncRNAs) could be used as biomarkers in the immune landscape of multiple tumors. In this study, we constructed a novel immune-related lncRNA (irlncRNA) risk model to predict the survival and immune landscape of GC patients. First, we identified differentially expressed irlncRNAs (DEirlncRNAs) from RNA-Seq data of The Cancer Genome Atlas (TCGA). By using various algorithms, we constructed a risk model with 11 DEirlncRNA pairs. We then tested the accuracy of the risk model, demonstrating that the risk model has good efficiency in predicting the prognosis of GC patients. Inner validation sets were further used to confirm the effectiveness of the risk model. In addition, our risk model has a preferable performance in predicting the immune infiltration status of tumors, immune checkpoint status of the patients, and immunotherapy score. In conclusion, our risk model may provide insights into the prognosis of and immunotherapy strategy for GC.

Development of a bioinformatics platform for analysis of quantitative transcriptomics and proteomics data: the OMnalysis

BACKGROUND

In the past decade, RNA sequencing and mass spectrometry based quantitative approaches are being used commonly to identify the differentially expressed biomarkers in different biological conditions. Data generated from these approaches come in different sizes (e.g., count matrix, normalized list of differentially expressed biomarkers, etc.) and shapes (e.g., sequences, spectral data, etc.). The list of differentially expressed biomarkers is used for functional interpretation and retrieve biological meaning, however, it requires moderate computational skills. Thus, researchers with no programming expertise find difficulty in data interpretation. Several bioinformatics tools are available to analyze such data; however, they are less flexible for performing the multiple steps of visualization and functional interpretation.

IMPLEMENTATION

We developed an easy-to-use Shiny based web application (named as OMnalysis) that provides users with a single platform to analyze and visualize the differentially expressed data. The OMnalysis accepts the data in tabular form from edgeR, DESeq2, MaxQuant Perseus, R packages, and other similar software, which typically contains the list of differentially expressed genes or proteins, log of the fold change, log of the count per million, the P value, q-value, etc. The key features of the OMnalysis are multiple image type visualization and their dimension customization options, seven multiple hypothesis testing correction methods to get more significant gene ontology, network topology-based pathway analysis, and multiple databases support (KEGG, Reactome, PANTHER, biocarta, NCI-Nature Pathway Interaction Database PharmGKB and STRINGdb) for extensive pathway enrichment analysis. OMnalysis also fetches the literature information from PubMed to provide supportive evidence to the biomarkers identified in the analysis. In a nutshell, we present the OMnalysis as a well-organized user interface, supported by peer-reviewed R packages with updated databases for quick interpretation of the differential transcriptomics and proteomics data to biological meaning.

AVAILABILITY

The OMnalysis codes are entirely written in R language and freely available at https://github.com/Punit201016/OMnalysis. OMnalysis can also be accessed from - http://lbmi.uvlf.sk/omnalysis.html. OMnalysis is hosted on a Shiny server at https://omnalysis.shinyapps.io/OMnalysis/. The minimum system requirements are: 4 gigabytes of RAM, i3 processor (or equivalent). It is compatible with any operating system (windows, Linux or Mac). The OMnalysis is heavily tested on Chrome web browsers; thus, Chrome is the preferred browser. OMnalysis works on Firefox and Safari.

Carceral-community epidemiology, structural racism, and COVID-19 disparities

Black and Hispanic communities are disproportionately affected by both incarceration and COVID-19. The epidemiological relationship between carceral facilities and community health during the COVID-19 pandemic, however, remains largely unexamined. Using data from Cook County Jail, we examine temporal patterns in the relationship between jail cycling (i.e., arrest and processing of individuals through jails before release) and community cases of COVID-19 in Chicago ZIP codes. We use multivariate regression analyses and a machine-learning tool, elastic regression, with 1,706 demographic control variables. We find that for each arrested individual cycled through Cook County Jail in March 2020, five additional cases of COVID-19 in their ZIP code of residence are independently attributable to the jail as of August. A total 86% of this additional disease burden is borne by majority-Black and/or -Hispanic ZIPs, accounting for 17% of cumulative COVID-19 cases in these ZIPs, 6% in majority-White ZIPs, and 13% across all ZIPs. Jail cycling in March alone can independently account for 21% of racial COVID-19 disparities in Chicago as of August 2020. Relative to all demographic variables in our analysis, jail cycling is the strongest predictor of COVID-19 rates, considerably exceeding poverty, race, and population density, for example. Arrest and incarceration policies appear to be increasing COVID-19 incidence in communities. Our data suggest that jails function as infectious disease multipliers and epidemiological pumps that are especially affecting marginalized communities. Given disproportionate policing and incarceration of racialized residents nationally, the criminal punishment system may explain a large proportion of racial COVID-19 disparities noted across the United States.

3D Spheroids Versus 3D Tumor-Like Microcapsules: Confinement and Mechanical Stress May Lead to the Expression of Malignant Responses in Cancer Cells

As 2D surfaces fail to resemble the tumoral milieu, current discussions are focused on which 3D cell culture strategy may better lead the cells to express in vitro most of the malignant hints described in vivo. In this study, this question is assessed by analyzing the full genetic profile of MCF7 cells cultured either as 3D spheroids-considered as "gold standard" for in vitro cancer research- or immobilized in 3D tumor-like microcapsules, by RNA-Seq and transcriptomic methods, allowing to discriminate at big-data scale, which in vitro strategy can better resemble most of the malignant features described in neoplastic diseases. The results clearly show that mechanical stress, rather than 3D morphology only, stimulates most of the biological processes involved in cancer pathogenicity, such as cytoskeletal organization, migration, and stemness. Furthermore, cells entrapped in hydrogel-based scaffolds are likely expressing other physiological hints described in malignancy, such as the upregulated expression of metalloproteinases or the resistance to anticancer drugs, among others. According to the knowledge, this study represents the first attempt to answer which 3D experimental system can better mimic the neoplastic architecture in vitro, emphasizing the relevance of confinement in cancer pathogenicity, which can be easily achieved by using hydrogel-based matrices.

12 Survival-related differentially expressed genes based on the TARGET-osteosarcoma database

The Therapeutically Applicable Research to Generate Effective Treatments (TARGET) project aims to determine molecular changes that drive childhood cancers, including osteosarcoma. The main purpose of the program is to use the open-source database to develop novel, effective, and less toxic therapies. We downloaded TARGET-OS RNA-Sequencing data through R studio and merged the mRNA expression of genes with clinical information (vital status, survival time and gender). Further, we analyzed differential gene expressions between dead and alive patients based on TARGET-OS project. By this study, we found 5758 differentially expressed genes between deceased and alive patients with a false discovery rate below 0.05; 4469 genes were upregulated in deceased patients compared to alive, whereas 1289 genes were downregulated. The survival-related genes were obtained using Kaplan-Meier survival analysis and Cox univariate regression (KM < 0.05 and Cox P-value < 0.05). Out of 5758 differentially expressed genes, only 217 have been associated with overall survival. Eight survival-related downregulated genes (ERCC4, CLUAP1, CTNNBIP1, GCA, RAB40C, SIRPA, USP11, and TCN2) and four survival-related upregulated genes (MUC1, COL13A1, JAG2 and KAZALD1) were selected for further analysis as potential independent prognostic candidate genes. This study may help to discover novel prognostic markers and potential therapeutic targets for osteosarcoma.

The role of mRNA in the development, diagnosis, treatment and prognosis of neural tumors

Neural tumors can generally be divided into central nervous system tumors and peripheral nervous tumors. Because this type of tumor is located in the nerve, even benign tumors are often difficult to remove by surgery. In addition, the majority of neural tumors are malignant, and it is particular the same for the central nervous system tumors. Even treated with the means such as chemotherapy and radiotherapy, they are also difficult to completely cure. In recent years, an increasingly number of studies have focused on the use of mRNA to treat tumors, representing an emerging gene therapy. The use of mRNA can use the expression of some functional proteins for the treatment of genetic disorders or tissue repair, and it can also be applied to immunotherapy through the expression of antigens, antibodies or receptors. Therefore, although these therapies are not fully-fledged enough, they have a broad research prospect. In addition, there are many ways to treat tumors using mRNA vaccines and exosomes carrying mRNA, which have drawn much attention. In this study, we reviewed the current research on the role of mRNA in the development, diagnosis, treatment and prognosis of neural tumors, and examine the future research prospects of mRNA in neural tumors and the opportunities and challenges that will arise in the future application of clinical treatment.

Assessing the risk of dengue severity using demographic information and laboratory test results with machine learning

BACKGROUND

Dengue virus causes a wide spectrum of disease, which ranges from subclinical disease to severe dengue shock syndrome. However, estimating the risk of severe outcomes using clinical presentation or laboratory test results for rapid patient triage remains a challenge. Here, we aimed to develop prognostic models for severe dengue using machine learning, according to demographic information and clinical laboratory data of patients with dengue.

METHODOLOGY/PRINCIPAL FINDINGS

Out of 1,581 patients in the National Cheng Kung University Hospital with suspected dengue infections and subjected to NS1 antigen, IgM and IgG, and qRT-PCR tests, 798 patients including 138 severe cases were enrolled in the study. The primary target outcome was severe dengue. Machine learning models were trained and tested using the patient dataset that included demographic information and qualitative laboratory test results collected on day 1 when they sought medical advice. To develop prognostic models, we applied various machine learning methods, including logistic regression, random forest, gradient boosting machine, support vector classifier, and artificial neural network, and compared the performance of the methods. The artificial neural network showed the highest average discrimination area under the receiver operating characteristic curve (0.8324 ± 0.0268) and balance accuracy (0.7523 ± 0.0273). According to the model explainer that analyzed the contributions/co-contributions of the different factors, patient age and dengue NS1 antigenemia were the two most important risk factors associated with severe dengue. Additionally, co-existence of anti-dengue IgM and IgG in patients with dengue increased the probability of severe dengue.

CONCLUSIONS/SIGNIFICANCE

We developed prognostic models for the prediction of dengue severity in patients, using machine learning. The discriminative ability of the artificial neural network exhibited good performance for severe dengue prognosis. This model could help clinicians obtain a rapid prognosis during dengue outbreaks. However, the model requires further validation using external cohorts in future studies.

Performance-weighted-voting model: An ensemble machine learning method for cancer type classification using whole-exome sequencing mutation

Background

With improvements in next-generation DNA sequencing technology, lower cost is needed to collect genetic data. More machine learning techniques can be used to help with cancer analysis and diagnosis.

Methods

We developed an ensemble machine learning system named performance-weighted-voting model for cancer type classification in 6,249 samples across 14 cancer types. Our ensemble system consists of five weak classifiers (logistic regression, SVM, random forest, XGBoost and neural networks). We first used cross-validation to get the predicted results for the five classifiers. The weights of the five weak classifiers can be obtained based on their predictive performance by solving linear regression functions. The final predicted probability of the performance-weighted-voting model for a cancer type can be determined by the summation of each classifier's weight multiplied by its predicted probability.

Results

Using the somatic mutation count of each gene as the input feature, the overall accuracy of the performance-weighted-voting model reached 71.46%, which was significantly higher than the five weak classifiers and two other ensemble models: the hard-voting model and the soft-voting model. In addition, by analyzing the predictive pattern of the performance-weighted-voting model, we found that in most cancer types, higher tumor mutational burden can improve overall accuracy.

Conclusion

This study has important clinical significance for identifying the origin of cancer, especially for those where the primary cannot be determined. In addition, our model presents a good strategy for using ensemble systems for cancer type classification.

Application of Artificial Intelligence Technology in Oncology: Towards the Establishment of Precision Medicine

In recent years, advances in artificial intelligence (AI) technology have led to the rapid clinical implementation of devices with AI technology in the medical field. More than 60 AI-equipped medical devices have already been approved by the Food and Drug Administration (FDA) in the United States, and the active introduction of AI technology is considered to be an inevitable trend in the future of medicine. In the field of oncology, clinical applications of medical devices using AI technology are already underway, mainly in radiology, and AI technology is expected to be positioned as an important core technology. In particular, "precision medicine," a medical treatment that selects the most appropriate treatment for each patient based on a vast amount of medical data such as genome information, has become a worldwide trend; AI technology is expected to be utilized in the process of extracting truly useful information from a large amount of medical data and applying it to diagnosis and treatment. In this review, we would like to introduce the history of AI technology and the current state of medical AI, especially in the oncology field, as well as discuss the possibilities and challenges of AI technology in the medical field.

Circulating tumor cells as Trojan Horse for understanding, preventing, and treating cancer: a critical appraisal

Circulating tumor cells (CTCs) are regarded as harbingers of metastases. Their ability to predict response to therapy, relapse, and resistance to treatment has proposed their value as putative diagnostic and prognostic indicators. CTCs represent one of the zeniths of cancer evolution in terms of cell survival; however, the triggers of CTC generation, the identification of potentially metastatic CTCs, and the mechanisms contributing to their heterogeneity and aggressiveness represent issues not yet fully deciphered. Thus, prior to enabling liquid biopsy applications to reach clinical prime time, understanding how the above mechanistic information can be applied to improve treatment decisions is a key challenge. Here, we provide our perspective on how CTCs can provide mechanistic insights into tumor pathogenesis, as well as on CTC clinical value. In doing so, we aim to (a) describe how CTCs disseminate from the primary tumor, and their link to epithelial-mesenchymal transition (EMT); (b) trace the route of CTCs through the circulation, focusing on tumor self-seeding and the possibility of tertiary metastasis; (c) describe possible mechanisms underlying the enhanced metastatic potential of CTCs; (d) discuss how CTC could provide further information on the tissue of origin, especially in cancer of unknown primary origin. We also provide a comprehensive review of meta-analyses assessing the prognostic significance of CTCs, to highlight the emerging role of CTCs in clinical oncology. We also explore how cell-free circulating tumor DNA (ctDNA) analysis, using a combination of genomic and phylogenetic analysis, can offer insights into CTC biology, including our understanding of CTC heterogeneity and tumor evolution. Last, we discuss emerging technologies, such as high-throughput quantitative imaging, radiogenomics, machine learning approaches, and the emerging breath biopsy. These technologies could compliment CTC and ctDNA analyses, and they collectively represent major future steps in cancer detection, monitoring, and management.

Artificial intelligence (AI) and big data in cancer and precision oncology

Artificial intelligence (AI) and machine learning have significantly influenced many facets of the healthcare sector. Advancement in technology has paved the way for analysis of big datasets in a cost- and time-effective manner. Clinical oncology and research are reaping the benefits of AI. The burden of cancer is a global phenomenon. Efforts to reduce mortality rates requires early diagnosis for effective therapeutic interventions. However, metastatic and recurrent cancers evolve and acquire drug resistance. It is imperative to detect novel biomarkers that induce drug resistance and identify therapeutic targets to enhance treatment regimes. The introduction of the next generation sequencing (NGS) platforms address these demands, has revolutionised the future of precision oncology. NGS offers several clinical applications that are important for risk predictor, early detection of disease, diagnosis by sequencing and medical imaging, accurate prognosis, biomarker identification and identification of therapeutic targets for novel drug discovery. NGS generates large datasets that demand specialised bioinformatics resources to analyse the data that is relevant and clinically significant. Through these applications of AI, cancer diagnostics and prognostic prediction are enhanced with NGS and medical imaging that delivers high resolution images. Regardless of the improvements in technology, AI has some challenges and limitations, and the clinical application of NGS remains to be validated. By continuing to enhance the progression of innovation and technology, the future of AI and precision oncology show great promise.