Autosurv: interpretable deep learning framework for cancer survival analysis incorporating clinical and multi-omics data.

Accurate prognosis for cancer patients can provide critical information for optimizing treatment plans and improving life quality. Combining omics data and demographic/clinical information can offer a more comprehensive view of cancer prognosis than using omics or clinical data alone and can also reveal the underlying disease mechanisms at the molecular level. In this study, we developed and validated a deep learning framework to extract information from high-dimensional gene expression and miRNA expression data and conduct prognosis prediction for breast cancer and ovarian-cancer patients using multiple independent multi-omics datasets. Our model achieved significantly better prognosis prediction than the current machine learning and deep learning approaches in various settings. Moreover, an interpretation method was applied to tackle the "black-box" nature of deep neural networks and we identified features (i.e., genes, miRNA, demographic/clinical variables) that were important to distinguish predicted high- and low-risk patients. The significance of the identified features was partially supported by previous studies.

AUTOSURV: INTERPRETABLE DEEP LEARNING FRAMEWORK FOR CANCER SURVIVAL ANALYSIS INCORPORATING CLINICAL AND MULTI-OMICS DATA.

Accurate prognosis for cancer patients can provide critical information for optimizing treatment plans and improving life quality. Combining omics data and demographic/clinical information can offer a more comprehensive view of cancer prognosis than using omics or clinical data alone and can reveal the underlying disease mechanisms at the molecular level. In this study, we developed a novel deep learning framework to extract information from high-dimensional gene expression and miRNA expression data and conduct prognosis prediction for breast cancer and ovarian cancer patients. Our model achieved significantly better prognosis prediction than the conventional Cox Proportional Hazard model and other competitive deep learning approaches in various settings. Moreover, an interpretation approach was applied to tackle the "black-box" nature of deep neural networks and we identified features (i.e., genes, miRNA, demographic/clinical variables) that made important contributions to distinguishing predicted high- and low-risk patients. The identified associations were partially supported by previous studies.

Detection of AD-BMP-2-IRES-HIF-1α MU on local promoting angiogenic and osteogenic capacity of necrosis area.

The induced EPCs were transfected by Ad-BMP-2-IRES-HIF-1αmu, and then transplanted into femoral head necrotic zone, the effect on osteogenesis and agiogenesis of necrosis zone was detected. The Ad-BMP-2-IRES-HIF-1α was transfected into induced EPCs and then transplanted into avascular necrotic parts of the femoral head (ANFH).Afterwards, the promotion effect on angiogenic and osteogenic capabilities of the necrosis parts from Ad-BMP-2-IRES-HIF-1α was detected. Rabbit bone marrow MNCs were obtained by density gradient centrifugation method, and were induced into EPCs by M199 medium; EPCs were identified in accordance with the cell morphology, specific surface markers and uptake abilities. The Ad-BMP-2-IRES-HIF-1α was transfected to EPCs and then transplanted into parts of ANFH. The models were euthanized 2 and 4 weeks after operation and then the angiogenic and osteogenic indexes of necrotic parts were detected. The results showed that more blood vessels generated in group A than that in group B and C (P<0.05), and the statistical differences were found between group B and C (P<0.05). The detection of histology and BMP-2 immunohistochemistry showed that there were statistically significant differences between group A and B, group A and C (P<0.05). There was no significant difference between group B and C (P<0.05). To sum up, this experiment shows that the EPCs transfected by Ad-BMP-2-IRES-HIF-1α have stronger angiogenic and osteogenic capabilities.