| Background:Breast cancer is one of the most serious cancers in the world.The number of cancer patients diagnosed in 2020 will reach 19.3 million,and 10 million will die of cancer,according to Andre ilbavi,who is a cancer expert.At present,one in five people in the world will suffer from cancer in their lifetime,and breast cancer has become the most common cancer because of the number of new cancer cases reaching more than 1.2 million,accounting for 11.7%.The pathogenesis of breast cancer is often the result of multiple factors,among which polygene is an important and key factor.Therefore,the in-depth study of breast cancer-related genes not only needs to analyze the expression level from different levels,but also needs to combine with a lot of omics information,so as to deeply and comprehensively study the role of a gene in the occurrence and development of breast cancer,and then has great significance for the early diagnosis and treatment of breast cancer.Bioinformatics is a new discipline which collects and selects biological data,edits,arranges and utilizes biological big data.It integrates many disciplines such as molecular biology,statistics,medical informatics and computer science,and penetrates into each other.For example,according to nucleic acid sequencing technology to predict the occurrence and development of diseases;according to DNA sequence to predict protein sequence and its function.If bioinformatics technology is combined with gene chip technology to deeply study the genes related to tumor pathogenesis and their regulation,it will reflect the integrity and systematicness of the research on the regulation mechanism of tumor gene expression.Although this speculation can not achieve high accuracy at present,it is like a beacon light for our further experimental research.With the application of different high-throughput technologies in medical biology,imageomics emerges as the times require.Imaging genomics is a new discipline to study the relationship between imaging phenotype and tumor genome.Through the interactive fusion of imageomics data and molecular omics data,not only the relationship between molecules and images is established,but also the relationship between macro and micro is established.The cancer imaging Archive(TCIA)is an important and reliable resource to realize the above Association.If the cancer imaging database is used to establish the relationship between macro imaging features and micro molecules,the targeted treatment of cancer will enter a new stage.Objective:The purpose of this study is to explore the possible role of ALPL gene in the occurrence of breast cancer by using multiple sets of information,to reveal the relationship between ALPL and calcification of breast cancer,and to provide a new idea for exploring the pathogenesis of breast cancer,and to lay a foundation for precise molecular diagnosis and targeted therapy of breast cancer.Methods and results:In this study,we first downloaded the difference of ALPL gene expression in different tumor tissues and breast cancer tissues through oncomine,a large tumor database,and then confirmed it through CCLE,a tumor cell line database.It was found that the difference of ALPL gene expression between breast cancer cases and normal people was significant.Then,we combined with the bioinformatics tool GSEA software and data mining methods to enrich and analyze the ALPL gene and the gene set that interacts with it,so as to predict the signal pathway of ALPL gene involved in the pathogenesis and its possible role in the occurrence of breast cancer.In addition,we used the human protein atlas database HPA to explore the differential expression of ALPL in breast cancer from the protein level,and verified it by immunohistochemistry In addition,we also used the string protein database to further analyze ALPL protein and its interacting proteins,revealing the relationship between Alpl and calcification of breast cancer,providing a new idea for the exploration of the pathogenesis of breast cancer,and then for the accurate molecular diagnosis of breast cancer And targeted therapy.Through the immunohistochemical verification of ALPL expression in 146 breast cancer patients,we found that ALPL expression in breast cancer patients was significantly higher than that in normal tissues adjacent to the cancer,and we found that there was a significant correlation between ALPL expression and HER-2 positive by analyzing the relationship between ALPL expression and HER-2 positive.In addition,we also divided 99 breast cancer images into high expression ALPL group(50 cases)and low expression ALPL group(49 cases).Using 3D slicer for image feature analysis,we established four machine learning algorithm models for judging ALPL expression: decision tree,extreme gradient lifting,support vector machine and neural network,and verified the confusion matrix and subject characteristic curve(ROC).Moreover,we found that the combination of ALPL gene expression and imaging characteristic parameters can improve the ability of imaging to judge the expression of CD8 a gene in breast cancer rate.Conclusion:This study comprehensively analyzed and predicted the role of ALPL gene in the occurrence of breast cancer by using multiomics.Combined with published studies,the increased alkaline phospholipase activity in breast cancer cases with ALPL overexpression may be closely related to the calcification sign of breast cancer,and this connection may be related to the overexpression of HER-2.In view of the close relationship between calcification sign and prognosis of breast cancer,ALPL is likely to be a highly effective biomarker for prognosis of breast cancer.Imaging omics has the potential to judge the gene expression status and tumor molecular subtypes of breast cancer.In the future,we will further study the effect of ALPL combined with image genomics on the diagnosis and typing of breast cancer by expanding the number of cases,so as to form a molecular marker network together with other molecular markers of breast cancer. |
PDF Full Text Request