Multiple myeloma (MM) remains an incurable disease primarily due to the emergence of drug resistance, and the underlying mechanisms remain unclear. Extrachromosomal circular DNAs (eccDNAs) are prevalent in cancer genomes of both coding and non‐coding regions. However, the role of non‐coding eccDNA regions that serve as enhancers has been largely overlooked. Here, genome‐wide profiling of serum eccDNAs from donors and MM patients who responded well or poorly to bortezomib‐lenalidomide‐dexamethasone (VRd) therapy is characterized. A high copy number of eccDNA ANKRD28 (eccANKRD28) predicts poor therapy response and prognosis but enhanced transcriptional activity. Established VRd‐resistant MM cell lines exhibit a higher abundance of eccANKRD28, and CRISPR/Cas9‐mediated elevation of eccANKRD28 desensitizes bortezomib and lenalidomide treatment both in vitro and in vivo. Integrated multi‐omics analysis (H3K27ac ChIP‐seq, scRNA‐seq, scATAC‐seq, CUT&Tag, et al.) identifies eccANKRD28 as an active enhancer involved in drug resistance driven by the key transcription factor, POU class 2 homeobox 2 (POU2F2). POU2F2 interacts with sequence‐specific eccANKRD28 as well as RUNX1 and RUNX2 motifs to form the protein complex, which activates the promoter of oncogenes, including IRF4, JUNB, IKZF3, RUNX3, and BCL2. This study elucidates the potential transcriptional network of enhancer eccANKRD28 in MM drug resistance from a previously unrecognized epigenetic perspective.
With the interconnection of distributed generation (DG), the new problems have been brought to traditional protections, such as sensitivity decrease, mal-operation and reject-operation, which deeply affects system operating and controlling. As the most troublesome failure of relay protection, hidden failure needs more attention. An analytic model based on protections and breakers signals is proposed to detect hidden failures in this paper. As a result of the uncertainty of signal transferring and equipment acting, there has multiple solution problem in the model. In order to solve this problem, an effective method was proposed by building the decision risk index. A stochastic optimization algorithm was employed to solve this model with consideration of a priori knowledge about the probabilities of the uncertain elements mentioned above. An example demonstrates that this method can be used to check the hidden failure effectively.