Data Sciences

Data Scientists at Rancho BioSciences are experts in large scale automated curation, annotation, and visualization. We build custom programmatic pipelines for comprehensive data cleanup and analysis that allow for complete reproducibility, facile incorporation of existing workflows, and interaction with third party platforms.

Large scale harmonization of clinical, molecular and public domain data
  • Rancho scientists lead the data management and integration efforts for several international consortium projects. This required integration of diverse clinical cohort data, large scale data transfer and organization, and preparation of combined meta-cohort statistical summaries.
  • Rancho supported supervised machine learning annotation for the IBM Watson AI platform to enable the discovery of new biological connections, novel drug targets, and new indications for existing drugs.
Software development for specific analysis
  • Rancho improved the protein array normalization, M statistic tests, batch correction, and heatmap functions in the PAA (Protein Array Analyzer) Bioconductor R package. These improvements were included in the April 2016 Bioconductor release. The modified package, complete with examples, is also available on GitHub. (
  • Rancho data scientists created a custom data curation package for extracting clinical data from trial report spreadsheets, which allowed our client to curate clinical oncology terms, export to an integrated database, and validate the results as well as ensured reproducibility.
  • Illumina BeadArray Data Processing – raw data to publication-quality figures in 10 clicks. Rancho bioinformatics scientists have expertise in constructing complete analysis pipelines that rival commercial software in performance and features, yet far exceed it in simplicity, flexibility, and, ultimately, value.
  • Rancho staff have adapted R scripts for various third party analysis platforms. For example, a tumor growth script, which uses tumor size and body weight measurements from animal studies to evaluate the efficacy of experimental and standard-of-care drug treatments, was integrated into tranSMART (
Interactive data visualization with custom Shiny Applications
  • Rancho scientists have developed custom visualization applications using the R programming language. These have enabled our clients to filter big data resources, perform ad-hoc statistical tests, incorporate new results or public data annotations, and export summary tables and graphics for presentations – without writing any code.

Rancho BioSciences is an international company and we provide data curation services in the following languages: English / French / German / Chinese / Spanish.

Please call (760) 717-7881 or email to discuss your specific project and needs.