HotSPOT: A Computational Tool to Design Targeted Sequencing Panels to Assess Early Photocarcinogenesis

Abstract

Mutations found in skin are acquired in specific patterns, clustering around mutation-prone genomic locations. The most mutation-prone genomic areas, mutation hotspots, first induce the growth of small cell clones in healthy skin. Mutations accumulate over time, and clones with driver mutations may give rise to skin cancer. Early mutation accumulation is a crucial first step in photocarcinogenesis. Therefore, a sufficient understanding of the process may help predict disease onset and identify avenues for skin cancer prevention. Early epidermal mutation profiles are typically established using high-depth targeted next-generation sequencing. However, there is currently a lack of tools for designing custom panels to capture mutation-enriched genomic regions efficiently. To address this issue, we created a computational algorithm that implements a pseudo-exhaustive approach to identify the best genomic areas to target. We benchmarked the current algorithm in three independent mutation datasets of human epidermal samples. Compared to the sequencing panel designs originally used in these publications, the mutation capture efficacy (number of mutations/base pairs sequenced) of our designed panel improved 9.6–12.1-fold. Mutation burden in the chronically sun-exposed and intermittently sun-exposed normal epidermis was measured within genomic regions identified by hotSPOT based on cutaneous squamous cell carcinoma (cSCC) mutation patterns. We found a significant increase in mutation capture efficacy and mutation burden in cSCC hotspots in chronically sun-exposed vs. intermittently sun-exposed epidermis (p < 0.0001). Our results show that our hotSPOT web application provides a publicly available resource for researchers to design custom panels, enabling efficient detection of somatic mutations in clinically normal tissues and other similar targeted sequencing studies. Moreover, hotSPOT also enables the comparison of mutation burden between normal tissues and cancer.

Mutation Hotspots Found in Bladder Cancer Aid Prediction of Carcinogenic Risk in Normal Urothelium

“…Therefore, we have created the hotSPOT tool to identify optimal genomic targets for studying mutations in normal tissues. These studies demonstrate the ability of the hotSPOT tool [20] to identify highly mutated genomic regions among bladder cancer samples. All identified hotspots in NMIBC were also found to be present in MIBC, demonstrating the accumulation of mutation burden during cancer progression.…”

“…We have developed a software tool, hotSPOT, that locates genomic regions with high mutation frequency [20]. Somatic mutation datasets for IMPACT targeted sequencing of 103 NMIBC samples (MSK Eur Urol 2017) [21], and whole-exome sequencing of 409 MIBC samples (TCGA, 2017) [22] were run through hotSPOT as discovery cohorts to identify highly mutated genomic regions in both cancer sub-types.…”

“…We have developed an algorithm to design targeted sequencing panels based on genomic areas that harbor high-frequency mutations [20]. This algorithm, hotSPOT, uses SNV data from any sample type and discovers highly mutated areas specific to the inputted dataset.…”