top of page
  • vitod1

IMMUNOTAR - Integrative Prioritization of Surface Proteins from Proteogenomic Data to Identify Novel

Shraim, Rawan MS. 1,2 , Weiner, Amber K. Ph.D 1, Conkrite, Karina L. 1, Mooney, Brian Ph.D. 4 , Maris, John M. MD 1,3, Diskin, Sharon J. Ph.D 1,3 , and Sacan, Ahmet Ph.D 2 1 Division of Oncology and Center for Childhood Cancer Research, Children's Hospital of Philadelphia, Philadelphia, PA, USA 2 School of Biomedical Engineering, Science and Health System, Drexel University, Philadelphia, PA 19104, USA 3 Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA 4 Department of Molecular Oncology, BC Cancer Research Institute, Vancouver, BC, Canada

Poster # 70

Immunotherapy harnesses the power of the immune system to combat diseases. Antibody drug conjugates and chimeric-antigen receptor T-cell therapy have shown advances in treating malignant tumors. Amid the progress, an obstacle is to identify cancer specific surface antigens. Here, we developed a computational tool, IMMUNOTAR, that integrates proteogenomic datasets to identify cancer specific surface protein immunotherapeutic targets. IMMUNOTAR combines and integrates quantitative attributes from user-provided RNA or protein cancer expression data with public proteogenomic datasets including: healthy-tissue RNA-sequencing and proteomics from GTEx, Evo-Devo-mammalian-organs and available publications, functional annotations from Gene Ontology, the pediatric FDA relevant molecular targets list, gene dependencies in cancer phenotypes from DepMap, and protein localization information from Uniprot, CIRFESS and Compartments. The user may choose the public datasets to enrich with, normalization and scaling methods, how to handle missing values, and assign prioritization weights to the attributes. The attributes for each gene are combined using a weighted linear summation to obtain the prioritization score. The overall performance of the tool is assessed based on the prioritization rankings of known therapeutic targets using mean-average-precision (MAP). Weight parameters are optimized using stochastic (SANN) and direct search (Nelder-Mead) algorithms to maximize the MAP score. The tool was tested on a cancer cell-line proteomics data from 8 pediatric cancers (Gonçalves et al. 2022). A total of 8,498 proteins were quantified and IMMUNOTAR identified an average of 47 differentially expressed plasma membrane proteins in each cancer that scored in the top 1% of all proteins. MUC16 was identified as a promising therapeutic target in all 8 cancer-phenotypes, validating prior observations (Aithal et al. 2019). In addition, CSPG4 was identified as a potential target in osteosarcoma, again confirming prior observations (Ricardo et al. 2019). In conclusion, IMMUNOTAR has shown the ability to identify currently sought after targets and potential novel targets for immunotherapy.

40 views0 comments

Recent Posts

See All


bottom of page