Background: Next-generation sequencing (NGS) is an important tool in clinical oncology as it enables personalized cancer treatment. However, not all patients respond to therapies selected based on sequence information, and NGS alone has not revealed the mechanism(s) underlying such resistance. Methods: Here, for 1417 tumor samples, we performed whole transcriptome sequencing to reveal tumor variant expression information. The latter studies focused on a clinically-oriented 50-gene panel, in order to examine transcriptomic silencing of putative driver alterations. Results: In this large-scale study, roughly 15% of the true somatic SNVs were not expressed as RNA, and 23% of patients had > 1 non-expressed SNV. SNV-bearing genes found to be consistently transcribed were TP53, PIK3CA, and KRAS, while those with lower rates of transcription were ALK, CSF1R, ERBB4, GNAS, KDR, and RET. In addition, we determined the frequency of tumor mutations being of germline rather than somatic origin and showed the feasibility of applying such an approach by analysis of 1879 tumor samples using whole exome tumor (somatic)/normal (germline) NGS; this analysis determined that, of the potential single nucleotide variants (SNVs) identified, 33.8% were of germline rather than somatic origin (after filtering through known germline variants). Conclusions: Both the frequency of silenced variant transcription and the risk of falsely identifying germline mutations as somatic/tumor-related variants are important methodological and biological phenomena, respectively, to be considered when analyzing tumor samples. In particular, transcriptomics are a critical adjunct to genomics when evaluating patient tumors for actionable alterations because without expression of the target proteins, there will likely be therapeutic resistance.