Multi-semantic feature fusion attention network for binary code similarity detection.

Binary code similarity detection (BCSD) plays a big role in the process of binary application security test. It can be applied in several fields, such as software plagiarism detection, malware analysis, vulnerability detection. Most research is based on recurrent neural networks, which is difficult to get the overall or long-distance semantic information of functions. Besides, exiting works simply extract high-level semantic features, lacking in-depth investigations on the potential mechanisms for fusing low-level and high-level semantic features. In this paper we propose a multi-semantic feature fusion attention network (MFFA-Net) for BCSD. MFFA-Net contains two critical modules: semantic feature fusion (SFF) and attention feature fusion (AFF). The SFF module concatenates multiple semantic features to represent the semantics of the function, which helps to obtain the overall semantic information of the function. The AFF module is designed to find useful information from various features, which assigns an attention matrix to research the relationship between features. In order to evaluate the proposed method, we made extensive experiments on two datasets. MFFA-Net can achieve a high degree of AUC at 99.6% and 98.3% respectively on the two datasets. The experimental results show that MFFA-Net has better performance for BCSD.

Identification of Pathogenic Genes and Transcription Factors in Osteosarcoma.

Osteosarcoma (OS) is an aggressive malignant tumor of the bones. Our study intended to identify and analyze potential pathogenic genes and upstream regulators for OS. We performed an integrated analysis to identify candidate pathogenic genes of OS by using three Gene Expression Omnibus (GEO) databases (GSE66673, GSE49003 and GSE37552). GO and KEGG enrichment analysis were utilized to predict the functional annotation and potential pathways of differentially expressed genes (DEGs). The OS-specific transcriptional regulatory network was established to study the crucial transcriptional factors (TFs) which target the DEGs in OS. From the three GEO datasets, we identified 759 DEGs between metastasis OS samples and non-metastasis OS samples. After GO and KEGG analysis, 'cell adhesion' (FDR = 1.27E-08), 'protein binding' (FDR = 1.13E-22), 'cytoplasm' (FDR = 5.63E-32) and 'osteoclast differentiation' (FDR = 0.000992221) were significantly enriched pathways for DEGs. HSP90AA1 exhibited a highest degree (degree = 32) and was enriched in 'pathways in cancer' and 'signal transduction'. BMP6, regulated by Pax-6, was enriched in the 'TGF-beta signaling pathway'. We indicated that BMP6 may be downregulated by Pax-6 in the non-metastasis OS samples. The up-regulated HSP90AA1 and down-regulated BMP6 and 'pathways in cancer' and 'signal transduction' were deduced to be involved in the pathogenesis of OS. The identified biomarkers and biological process in OS may provide foundation for further study.

Identification of gene biomarkers in patients with postmenopausal osteoporosis.

Postmenopausal osteoporosis (PMOP) is a major public health concern worldwide. The present study aimed to provide evidence to assist in the development of specific novel biomarkers for PMOP. Differentially expressed genes (DEGs) were identified between PMOP and normal controls by integrated microarray analyses of the Gene Expression Omnibus (GEO) database, and the optimal diagnostic gene biomarkers for PMOP were identified with LASSO and Boruta algorithms. Classification models, including support vector machine (SVM), decision tree and random forests models, were established to test the diagnostic value of identified gene biomarkers for PMOP. Functional annotations and protein­protein interaction (PPI) network constructions were also conducted. Integrated microarray analyses (GSE56815, GSE13850 and GSE7429) of the GEO database were employed, and 1,320 DEGs were identified between PMOP and normal controls. An 11­gene combination was also identified as an optimal biomarker for PMOP by feature selection and classification methods using SVM, decision tree and random forest models. This combination was comprised of the following genes: Dehydrogenase E1 and transketolase domain containing 1 (DHTKD1), osteoclast stimulating factor 1 (OSTF1), G protein­coupled receptor 116 (GPR116), BCL2 interacting killer, adrenoceptor ß1 (ADRB1), neogenin 1 (NEO1), RB binding protein 4 (RBBP4), GPR87, cylicin 2, EF­hand calcium binding domain 1 and DEAH­box helicase 35. RBBP4 (degree=12) was revealed to be the hub gene of this PMOP­specific PPI network. Among these 11 genes, three genes (OSTF1, ADRB1 and NEO1) were speculated to serve roles in PMOP by regulating the balance between bone formation and bone resorption, while two genes (GPR87 and GPR116) may be involved in PMOP by regulating the nuclear factor­κB signaling pathway. Furthermore, DHTKD1 and RBBP4 may be involved in PMOP by regulating mitochondrial dysfunction and interacting with ESR1, respectively. In conclusion, the findings of the current study provided an insight for exploring the mechanism and developing novel biomarkers for PMOP. Further studies are required to test the diagnostic value for PMOP prior to use in a clinical setting.

Identification of clinical tumor stages related mRNAs and miRNAs in cervical squamous cell carcinoma.

OBJECTIVES: The aim of this study is to identify the clinical tumor stage related mRNAs and miRNAs, shedding light on the potential molecular mechanisms of cervical squamous cell carcinoma (CSCC). METHODS: Firstly, the mRNA and miRNA next-generation sequencing data were downloaded. Secondly, clinical tumor stage correlation analysis of mRNAs and miRNA was performed, followed by the functional enrichment analysis of all clinical tumor stage related mRNAs. Thirdly, differentially expression analysis of mRNAs and miRNA between different clinical tumor stages was performed, followed by target gene prediction of these differentially expressed miRNAs. RESULTS: 3 mRNAs (PER1, PRKAB1 and PMM2) and 5 miRNAs (hsa-mir-486, hsa-mir-451, hsa-mir-424, hsa-mir-144 and hsa-mir-450a-2) were overlapped from stage 1, stage 2, stage 3 and stage 4. CONCLUSIONS: Alterations of differentially expressed mRNAs and miRNAs may offer important insights into the molecular mechanisms in the pathology of CSCC.