Characterization of the transcriptional regulatory network of the normal cell cycle is essential for understanding the perturbations that lead to cancer. However, the complete set of cycling genes in primary cells has not yet been identified. Here, we report the results of genome-wide expression profiling experiments on synchronized primary human foreskin fibroblasts across the cell cycle. Using a combined experimental and computational approach to deconvolve measured expression values into ''single-cell'' expression profiles, we were able to overcome the limitations inherent in synchronizing nontransformed mammalian cells. This allowed us to identify 480 periodically expressed genes in primary human foreskin fibroblasts. Analysis of the reconstructed primary cell profiles and comparison with published expression datasets from synchronized transformed cells reveals a large number of genes that cycle exclusively in primary cells. This conclusion was supported by both bioinformatic analysis and experiments performed on other cell types. We suggest that this approach will help pinpoint genetic elements contributing to normal cell growth and cellular transformation.deconvolution Í expression profile T ight regulation of the cell cycle is necessary for the proper growth and development of all organisms. Dysregulation of cell cycle controls leads to proliferative diseases, most notably cancer. One approach to understanding basic cell cycle processes and their deregulation in cancer has been genome-wide characterization of the cell cycle transcriptional program (1). In these microarray experiments, the RNA levels of every gene is measured in a synchronized cell population at multiple time points. Synchronization is achieved by releasing cells from a cell cycle arrest. This approach was carried out initially to characterize the yeast cell cycle, and, subsequently, it was applied to examine the cell cycle in multiple organisms (reviewed in ref. 2).Although arrest methods were effective for characterizing cycling genes in a number of species (3-7), they did not lead to complete synchronization, even for yeast cells (8)(9)(10). A number of methods were introduced for resynchronizing yeast cells by either matching the profiles for the first and second cycle for each gene (9) or by combining expression and bud count information to reconstruct the expression profile (8). These methods were shown to improve (the already good) yeast cell cycle expression data. However, these methods cannot be directly applied to mammalian cells because of two major differences between yeast and mammalian cells: (i) normal diploid mammalian cells lose their synchronization relatively soon after release of growth arrest (11) and (ii) only 50-70% of wild-type mammalian cells reenter the cell cycle after release from arrest (12). The large percentage of arrested cells and loss of synchronization means that expression values represent a mixed population of cells, which introduces high background noise that confounds differentiation between genuine c...