PT - JOURNAL ARTICLE AU - Anupama Rajan Bhat AU - Manoj Kumar Gupta AU - Priya Krithivasan AU - Kunal Dhas AU - Jayalakshmi Nair AU - Ram Bhupal Reddy AU - HV Sudheendra AU - Sandip Chavan AU - Harsha Vardhan AU - Sujatha Darsi AU - Lavanya Balakrishnan AU - Shanmukh Katragadda AU - Vikram Kekatpure AU - Amritha Suresh AU - Pramila Tata AU - Binay Panda AU - Moni A Kuriakose AU - Ravi Sirdeshmukh TI - An integrated transcriptomics and proteomics study of Head and Neck Squamous Cell Carcinoma – methodological and analytical considerations AID - 10.1101/024059 DP - 2015 Jan 01 TA - bioRxiv PG - 024059 4099 - http://biorxiv.org/content/early/2015/08/06/024059.short 4100 - http://biorxiv.org/content/early/2015/08/06/024059.full AB - High throughput molecular profiling and integrated data analysis with tumor tissues require overcoming challenges like tumor heterogeneity and tissue paucity. This study is an attempt to understand and optimize various steps during tissue processing and in establishing pipelines essential for integrated analysis. Towards this effort, we subjected laryngo-pharyngeal primary tumors and the corresponding adjacent normal tissues (n=2) to two RNA and protein isolation methods, one wherein RNA and protein were isolated from the same tissue sequentially (Method 1) and second, wherein the extraction was carried out using two independent methods (Method 2). RNA and protein from both methods were subjected to RNA-seq and iTRAQ based LC-MS/MS analysis. Transcript and peptide identification and quantification was followed by both individual-ome and integrated data analysis. As a result of this analysis, we identified a higher number of total, as well as differentially expressed (DE) transcripts (1329 vs 1134) and proteins (799 vs 408) with fold change ≥ 2.0, in Method 1. Among these, 173 and 86 entities were identified by both transcriptome and proteome analysis in Method 1 and 2, respectively, with higher concordance in the regulation trends observed in the former. The significant cancer related pathways enriched with the individual DE transcript or protein data were similar in both the methods. However, the entities mapping to them were different, allowing enhanced view of the pathways identified after integration of the data and subsequent mapping. The concordant DE transcripts and proteins also revealed key molecules of the pathways with important roles in cancer development. This study thus demonstrates that sequential extraction of the RNA and proteins from the same tissue allows for better profiling of differentially expressed entities and a more accurate integrated data analysis.Author Contributions ARB, MKG, PK and SK contributed final data analysis. KD and JN were involved in the RNASeq experiments while MKG, SHV LB and SC were involved in the iTRAQ MS/MS analysis. RBR and HV contributed towards the standardization of sample collection and processing, and were also involved in obtaining clinical information of the patients along with SD. VK and MAK were involved in study design, providing clinical insights into the analysis and in critical assessment of the manuscript. ARB, MKG and PK were involved in manuscript preparation. AS, PT, BP, MAK and RS were involved in the establishing the study design, overall monitoring of the experimental results and manuscript preparation. PT, MAK, BP and RS are the lead investigators of the project.Significance of the study The study highlights the need to optimize tissue processing and analytical pipelines to enable accurate integrated analysis of high throughput omics data; a sequential extraction of RNA and protein entities and subsequent integrated analysis was identified to provide a better representation of the molecular profile in terms concordant entities and pathways.