Mechanistic models of signaling pathways deconvolute the functional landscape of glioblastoma at single cell resolution

The rapid development of single cell RNA-sequencing (scRNA-seq) technologies is revealing an unexpectedly large degree of heterogeneity in gene expression levels across the different cells that compose the same tissue sample. However, little is known on the functional consequences of this heterogeneity and the contribution of individual cell-fate decisions to the collective behavior of the tissues these cells are part of. Mechanistic models of signaling pathways have already proven to be useful tools for understanding relevant aspects of cell functionality. Here we propose to use this mechanistic modeling strategy to deconvolute the complexity of the functional behavior of a tissue by dissecting it into the individual functional landscapes of its component cells by using a single-cell RNA-seq experiment of glioblastoma cells. This mechanistic modeling analysis revealed a high degree of heterogeneity at the scale of signaling circuits, suggesting the existence of a complex functional landscape at single cell level. Different clusters of neoplastic glioblastoma cells have been characterized according to their differences in signaling circuit activity profiles, which only partly overlap with the conventional glioblastoma subtype classification. The activity of signaling circuits that trigger cell functionalities which can easily be assimilated to cancer hallmarks reveals different functional strategies with different degrees of aggressiveness followed by any of the clusters. In addition, mechanistic modeling allows simulating the effect of interventions on the components of the signaling circuits, such as drug inhibitions. Thus, effects of drug inhibitions at single cell level can be dissected, revealing for the first time the mechanisms that individual cells use to avoid the effect of a targeted therapy which explain why and how a small proportion of cells display, in fact, different degrees of resistance to the treatment. The results presented here strongly suggest that mechanistic modeling at single cell level not only allows uncovering the molecular mechanisms of the tumor progression but also can predict the success of a treatment and can contribute to a better definition of therapeutic targets in the future.