Background The recent Coronavirus Disease 2019 (COVID-19) pandemic has placed severe stress on healthcare systems worldwide, which is amplified by the critical shortage of COVID-19 tests. Methods In this study, we propose to generate a more accurate diagnosis model of COVID-19 based on patient symptoms and routine test results by applying machine learning to reanalyzing COVID-19 data from 151 published studies. We aim to investigate correlations between clinical variables, cluster COVID-19 patients into subtypes, and generate a computational classification model for discriminating between COVID-19 patients and influenza patients based on clinical variables alone. Results We discovered several novel associations between clinical variables, including correlations between being male and having higher levels of serum lymphocytes and neutrophils. We found that COVID-19 patients could be clustered into subtypes based on serum levels of immune cells, gender, and reported symptoms. Finally, we trained an XGBoost model to achieve a sensitivity of 92.5% and a specificity of 97.9% in discriminating COVID-19 patients from influenza patients. Conclusions We demonstrated that computational methods trained on large clinical datasets could yield ever more accurate COVID-19 diagnostic models to mitigate the impact of lack of testing. We also presented previously unknown COVID-19 clinical variable correlations and clinical subgroups.

An intra-pancreatic microbiota was recently discovered in several prominent studies. Since pancreatic adenocarcinoma (PAAD) is one of the most lethal cancers worldwide, and the intratumor microbiome was found to be a significant contributor to carcinogenesis in other cancers, this study aims to characterize the PAAD microbiome and elucidate how it may be associated with PAAD prognosis. We further explored the association between the intra-pancreatic microbiome and smoking and gender, which are both risk factors for PAAD. RNA-sequencing data from The Cancer Genome Atlas (TCGA) were used to infer microbial abundance, which was correlated to clinical variables and to cancer and immune-associated gene expression, to determine how microbes may contribute to cancer progression. We discovered that the presence of several bacteria species within PAAD tumors is linked to metastasis and immune suppression. This is the first large-scale study to report microbiome-immune correlations in human pancreatic cancer samples. Furthermore, we found that the increased prevalence and poorer prognosis of PAAD in males and smokers are linked to the presence of potentially cancer-promoting or immune-inhibiting microbes. Further study into the roles of these microbes in PAAD is imperative for understanding how a pro-tumor microenvironment may be treated to limit cancer progression.

Although 1 in 9 American men will receive a diagnosis of prostate cancer (PC), most men with this diagnosis will not die from it, as most PCs are indolent. However, there is a subset of patients in which the once-indolent PC becomes metastatic and eventually, fatal. In this study, we analyzed microbial compositions of intratumor bacteria in PC to determine the influence of the microbiome on metastatic growth. Using large-scale RNA-sequencing data and corresponding clinical data, we correlated the abundance of microbes to immune pathways and PC risk factors, identifying specific microbes that either significantly deter or contribute to cancer aggressiveness. Interestingly, most of the microbes we found appeared to play anti-tumor roles in PC. Since these anti-tumor microbes were overrepresented in tumor samples, we believe that microbes thrive in the tumor microenvironment, outcompete cancer cells, and directly mitigate tumor growth by recruiting immune cells. These include Listeria monocytogenes, Methylobacterium radiotolerans JCM 2831, Xanthomonas albilineans GPE PC73, and Bradyrhizobium japonicum, which are negatively correlated with Gleason score, Tumor-Node-Metastasis (TNM) stage, prostate-specific antigen (PSA) level, and Androgen Receptor (AR) expression, respectively. We also identified microbes that contribute to tumor growth and are positively correlated with genomic alterations, dysregulated immune-associated (IA) genes, and prostate cancer stem cells (PCSC) genes.