SGCAL: An Algorithm to Identify Sensitive Gene Combinations in the Mouse Osteoblast Gene Network

Osteoblasts originate from mesenchymal stem cells (MSCs) and their maturation is regulated by numerous genes. However, specific genes that play decisive roles in osteoblast maturation are not completely understood. Herein, we propose a new algorithm called SGCAL to identify sensitive gene combinations during osteoblast development and decode the regulatory roles of sensitive gene combinations in cell development. In contrast to conventional gene identification algorithms, the SGCAL algorithm, which is based on a combination of ARNN-LNE iterative cycles, quantitatively evaluates the sensitivity of genes by analyzing the average network entropy before and after gene perturbation. We verified that osteoblast development is fundamentally regulated by an overall sensitive network formed by specific genes. Moreover, we explored the roles of these sensitive genes and their combinations in the transition time point of the initial state of network change. Our algorithm was validated on two independent mouse osteoblast development datasets, successfully identifying sensitive genes and their combinations in both datasets and revealing that gene combinations containing Col1a1 can exhibit significant sensitivity. This study demonstrated the universality of cell fate and network regulation, offering new insights and a theoretical basis for the treatment and repair of bone diseases