UALCAN: An update to the integrated cancer data analysis platform

Abstract

Cancer genomic, transcriptomic, and proteomic profiling has generated extensive data that necessitate the development of tools for its analysis and dissemination. We developed UALCAN to provide a portal for easy exploring, analyzing, and visualizing these data, allowing users to integrate the data to better understand the gene, proteins, and pathways perturbed in cancer and make discoveries. UALCAN web portal enables analyzing and delivering cancer transcriptome, proteomics, and patient survival data to the cancer research community. With data obtained from The Cancer Genome Atlas (TCGA) project, UALCAN has enabled users to evaluate protein-coding gene expression and its impact on patient survival across 33 types of cancers. The web portal has been used extensively since its release and received immense popularity, underlined by its usage from cancer researchers in more than 100 countries. The present manuscript highlights the task we have undertaken and updates that we have made to UALCAN since its release in 2017. Extensive user feedback motivated us to expand the resource by including data on a) microRNAs (miRNAs), long non-coding RNAs (lncRNAs), and promoter DNA methylation from TCGA and b) mass spectrometry-based proteomics from the Clinical Proteomic Tumor Analysis Consortium (CPTAC). UALCAN provides easy access to pre-computed, tumor subgroup-based gene/protein expression, promoter DNA methylation status, and Kaplan-Meier survival analyses. It also provides new visualization features to comprehend and integrate observations and aids in generating hypotheses for testing. UALCAN is accessible at http://ualcan.path.uab.edu.

Keywords: Bioinformatics; Cancer gene expression; Correlation; Differential expression; Patient survival; Proteomics; Transcriptomics; UALCAN; lncRNA.

Fig. 1

LncRNA expression profile in prostate adenocarcinoma. (A-D) Graphs showing expression level analysis of PCAT6 using UALCAN web-portal in normal prostate, all primary tumors and subgroups based on patient race (B), Gleason score(C) and molecular subtypes (D). (E) Pan cancer gene expression profile of PCAT6. Red boxplot depicts expression level in primary tumors, while blue boxplot indicate expression in normal samples.

Fig. 2

miRNA expression profile in Lung adenocarcinoma. (A) Heatmap generated using UALCAN showing top miRNAs under expressed in lung adenocarcinoma [LUAD]. (B-C) Boxplot showing expression level of hsa-miR-144 in normal, primary tumors and tumor subgroups based on patient's smoking status. (D) Predicted targets of hsa-miR-144 and their expression status is lung adenocarcinoma.

Fig. 3

Promoter methylation of DNA information from TCGA Illumina bead chip data in UALCAN. (A) UALCAN generated heatmap showing top 25 genes with hyper-methylated promoter DNA in breast invasive carcinoma. (B, C) Boxplots showing inverse relation between promoter methylation status and gene expression profile of ENPP2 in TCGA breast invasive carcinoma [BRCA]. (D-G) Boxplots showing promoter methylation level and mRNA expression level of ENPP2 in different stages of breast cancer (D,E) and different subclasses of breast cancer (F, G).

Fig. 4

Gene expression correlation analysis of FOXM1 in Breast invasive carcinoma [BRCA]. (A) Gene query result page in UALCAN directs user to gene correlation analysis page. (B, C) UALCAN generated list of positively [Pearson Correlation Coefficient > 0.3] and negatively [Pearson Correlation Coefficient < -0.3] correlated genes of FOXM1 in BRCA. Scatter plots are provided to visualize correlation for each gene pair.

Fig. 5

Total protein and phosphoprotein expression pattern of MUC1 in lung adenocarcinoma. (A) Boxplot generated using UALCAN showing total protein expression of MUC1 in lung adenocarcinoma. (B-D) Boxplots depicting expression level of MUC1 phosphoproteins in lung adenocarcinoma. (E) Pan cancer view of MUC1 total protein expression in breast, colon, ovarian, renal and endometrial cancer.