Gene panel sequencing data of 89 clear cell renal cell carcinoma patients (ccRCC)

We generated paired-end panel seq data from 89 ccRCC patients, including two spatially separated primary tumor biopsies, and a matched normal sample per patient. A selection of 826 genes frequently mutated in ccRCC was targeted at high coverage. The data was sequenced using the Illumina HiSeq 2500 system. The sequenced reads contain unique molecular identifiers (UMI), and this allows for the correction of potential sequencing errors. The UMIs are attached to the read id with an underscore. Reads with identical UMIs, which are mapped to the same genomic position, come from the same DNA molecule, and therefore, the consensus sequence can be built, and the variants can be called with higher confidence. The patients are labelled P01, P02, ... P89. From each, the normal sample has the label 'NO', the first sample from the primary tumor has the label 'TU1', and the second sample from the primary tumor has the label 'TU2'. Each sample was sequenced on multiple lanes. From the first 14 patients (P01 - P14), the tissue was frozen, while from the remaining patients (P15 - P89), the samples were FFPE. The sequencing kit was Agilent Haloplex HS. The adapter for read 1 was 'GAGATCGGAAGAGCACACGTCTGAACTCCAGTCA', and the adapter for read 2 was 'AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT'. Please note that, before processing the reads, the last base needs to be removed from the end of read 1, and the first base needs to be removed from the beginning of read 2. This is because the HaloPlexHS protocol use a vector that adds an extra base to the adapter.