Detection of COPB2 as a KRAS synthetic lethal partner through integration of functional genomics screens
Metrics: PDF 1430 views | HTML 2736 views | ?
Eleni G. Christodoulou1,2, Hai Yang3, Franziska Lademann4, Christian Pilarsky3,4, Andreas Beyer1,5, Michael Schroeder1
1Biotechnology Center, TU Dresden, Dresden, Germany
2Department of Medical Oncology, National Cancer Center of Singapore, Singapore
3Chirurgische Klinik, Translational Research Center, Universitätsklinikum Erlangen, Erlangen, Germany
4Medizinische Fakultät Carl Gustav Carus, TU Dresden, Dresden, Germany
5Cellular Networks and Systems Biology, University of Cologne, Cologne, Germany
Andreas Beyer, email: firstname.lastname@example.org
Michael Schroeder, email: email@example.com
Keywords: KRAS, synthetic lethal partner, ranking, rank aggregation, functional screens
Received: March 09, 2016 Accepted: March 01, 2017 Published: March 10, 2017
Mutated KRAS plays an important role in many cancers. Although targeting KRAS directly is difficult, indirect inactivation via synthetic lethal partners (SLPs) is promising. Yet to date, there are no SLPs from high-throughput RNAi screening, which are supported by multiple screens. Here, we address this problem by aggregating and ranking data over three independent high-throughput screens. We integrate rankings by minimizing the displacement and by considering established methods such as RIGER and RSA.
Our meta analysis reveals COPB2 as a potential SLP of KRAS with good support from all three screens. COPB2 is a coatomer subunit and its knock down has already been linked to disabled autophagy and reduced tumor growth. We confirm COPB2 as SLP in knock down experiments on pancreas and colorectal cancer cell lines.
Overall, consistent integration of high throughput data can generate candidate synthetic lethal partners, which individual screens do not uncover. Concretely, we reveal and confirm that COPB2 is a synthetic lethal partner of KRAS and hence a promising cancer target. Ligands inhibiting COPB2 may, therefore, be promising new cancer drugs.
All site content, except where otherwise noted, is licensed under a Creative Commons Attribution 3.0 License.