Manifold Peaks Nonnegative Matrix Factorization

Multi-Omics Analysis Elucidates The Immune And Intratumor Microbes Characteristics Of Ubiquitination Subtypes In Lung Adenocarcinoma

Ubiquitination modification is closely related to cancer and participates in the regulation of tumor microenvironment. However, the role of ubiquitination modification in the immune response and prognosis of lung adenocarcinoma has not been elucidated. This study aims to establish a disease classification associated with ubiquitination and reveal the landscape of intratumor microbes in patients with lung adenocarcinoma for the first time. A total of 1314 patients with lung adenocarcinoma in the GEO and TCGA databases were included in our study. We constructed a ubiquitination scoring model using WGCNA and constructed ubiquitination subtypes using unsupervised clustering, analyzed the clinical characteristics, immune characteristics, and intratumor microbes characteristics, and screened out the relevant gene signatures, which were verified by RT-qPCR in human cancer cells. The results showed that the high ubiquitination subtype had poor prognosis, low degree of immune infiltration, high index of tumor stemness, and poor effect of immunotherapy. The subtypes with lower ubiquitination scores have better prognosis, higher tumor microenvironment score and better immunotherapy effect. The C2 subtype has high level of immune infiltration, lower intratumor microbes diversity and abundance, and good prognosis. The C3 subtype has low level of immune infiltration, higher intratumor microbes diversity and abundance, and poor prognosis. The C1 subtype has characteristics between C2 and C3. In summary, this paper constructs a scoring system and several subtypes based on ubiquitination genes, and analyzed the characteristics, which can help provide new methods for clinical treatment.

Multiple Graph Adaptive Regularized Semi-Supervised Nonnegative Matrix Factorization with Sparse Constraint for Data Representation

Multiple graph and semi-supervision techniques have been successfully introduced into the nonnegative matrix factorization (NMF) model for taking full advantage of the manifold structure and priori information of data to capture excellent low-dimensional data representation. However, the existing methods do not consider the sparse constraint, which can enhance the local learning ability and improve the performance in practical applications. To overcome this limitation, a novel NMF-based data representation method, namely, the multiple graph adaptive regularized semi-supervised nonnegative matrix factorization with sparse constraint (MSNMFSC) is developed in this paper for obtaining the sparse and discriminative data representation and increasing the quality of decomposition of NMF. Particularly, based on the standard NMF, the proposed MSNMFSC method combines the multiple graph adaptive regularization, the limited supervised information and the sparse constraint together to learn the more discriminative parts-based data representation. Moreover, the convergence analysis of MSNMFSC is studied. Experiments are conducted on several practical image datasets in clustering tasks, and the clustering results have shown that MSNMFSC achieves better performance than several most related NMF-based methods.