IMCell predicts minimal, optimized transcription factor (TF) sets that drive cell identity transitions by formulating TF selection as an influence-maximization problem over a signed gene-regulatory network. IMCell jointly maximizes activation of target genes and repression of off-target genes using a greedy Monte-Carlo solver, supports optional expression-based node weighting, and can be extended (dynIMCell, via Epoch) to predict step-wise differentiation protocols. Software accompanying Su, Ly & Cahan, "Prediction of parsimonious and temporally-sensitive sets of cell fate engineering transcription factors with IMCell."