Abstract

SummaryData mining from RNA-seq or microarray data has become an essential part of cancer biomarker exploration. Certain existing web servers are valuable and broadly utilized, but the meta-analysis of multiple datasets is absent. Most web servers only contain tumor samples from the TCGA database with only one cohort for each cancer type, which also means that the analysis results mainly derived from a single cohort are thin and unstable. Indeed, consistent performance across multiple independent cohorts is the foundation for an excellent biomarker. Moreover, many analytical functions researchers require remain adequately unmet by these tools. Thus, we introduce BEST (Biomarker Exploration for Solid Tumors), a web application for comprehensive biomarker exploration on large-scale data in solid tumors. BEST includes more than 50,000 samples of 27 cancer types. To ensure the comparability of genes between different sequencing technologies and the legibility of clinical traits, we re-annotated transcriptome data based on the GRCh38 patch 13 sequences and unified the nomenclature of clinical traits. BEST delivers fast and customizable functions, including clinical association, survival analysis, enrichment analysis, cell infiltration, immunomodulator, immunotherapy, candidate agents, and genomic alteration. Together, our web server provides multiple cleaned-up independent datasets and diverse analysis functionalities, helping unleash the value of current data resources. It is freely available at https://rookieutopia.com/. The bigger pictureBioinformatics web servers enable researchers without computational programming skills to conduct various bioinformatics analyses. However, most web servers only contain tumor samples from the TCGA database with only one cohort for each cancer type, which also means that the analysis results mainly derived from a single cohort are thin and unstable. Thus, we introduce BEST (Biomarker Exploration for Solid Tumors), a web application for comprehensive biomarker exploration on large-scale data in solid tumors. BEST includes more than 50,000 samples of 27 cancer types that have been uniformly re-annotated based on the GRCh38 patch 13 sequences, which ensures the comparability of genes between different sequencing technologies. BEST also offers prevalent functions including clinical association, survival analysis, enrichment analysis, cell infiltration, immunomodulator, immunotherapy, candidate agents, and genomic alteration. Together, BEST provides a curated database and innovative analytical pipelines to explore cancer biomarkers at high resolution.