Implementation and use of cancer expression signatures from high resolution transcriptomics data

The employment of high-resolution transcriptomics is crucial to gain a more in-depth understanding of tumor biology, contributing to discovering new diagnostic and therapeutic insights in oncology. In particular, this kind of data allows for the analysis of individual cells separately, providing a more detailed picture of cellular heterogeneity within the tumor. Additionally, high-resolution transcriptomics enables the identification of crucial cellular subtypes in understanding tumors, as cellular diversity can impact disease progression and treatment response. This underscores the importance of discovering new tumor expression signatures to develop targeted and personalized therapies. Here, we present a collection of tumor signatures derived from high-resolution data and the method for computing their scores: a numerical assessment that identifies the presence or activity of a specific signature in a given biological sample. Subsequently, we integrated these signatures into "signifinder", an R package that streamlines the computation of signature scores from a normalized input dataset, making it faster and more user-friendly.