A Bayesian approach to determine the composition of heterogeneous cancer tissue

We tackle the challenge of determining the composition of a heterogeneous cancer tissue. There exist high cost methods providing cell level resolution to observe important feature values. This method can be used to get accurate estimates of hetergeneous cell composition but the cost is prohibitive. We propose an algorithm that can get accurate estimates of the heterogeneous cancer tissue composition in a cost effective manner. It relies on one-time high cost cell level resolution measurements which can then be utilized in low cost low resolution aggregate measurements to estimate the composition of the heterogeneous cancer tissue.