【摘 要】
：
Background: Since biological data are usually complex and high dimensional,selecting meaningful information from the biological data has been very important in biological data analysis.Methods: In org
【机 构】
：
计算机科学与技术学院,大连理工大学,大连,116024
【出 处】
：
第七届全国生物信息学与系统生物学学术大会
论文部分内容阅读
Background: Since biological data are usually complex and high dimensional,selecting meaningful information from the biological data has been very important in biological data analysis.Methods: In organisms,the physiological and pathological changes are usually influenced by molecule interactions.Hence,an interaction gain-recursive feature elimination (IG-RFE) method is proposed here to measure the feature importance based on the symmetric uncertainty between the feature and the class label and the interactions among the features.In each iteration,the correlation between the featurefand the class label is reflected by the symmetric uncertainty.The interaction of the featurefwith other features in the current feature set F is calculated by the average interaction gain[1]off,each g∈F-{f} and the class label.Based on the symmetric uncertainty and interaction gain,less important features are removed from the current feature set F in each loop.
其他文献
k-mer usage in genome sequences is not random,the nonrandom evolution and its cause are worthy of special attention.After analyzing 8-mer frequency spectrum in human,it is found that arisen frequency
Hepatocellular carcinoma (HCC) is one of the common malignant tumors.Accurate diagnosing of HCC is of great importance.k-NFN (k-Nearest and Farthest Neighbors)[1]is an ensemble classifying method whic
Schizophrenia and bipolar disorder are complex mental disorders,with risks contributed by multiple genes.Recently,genome-wide systemic approaches have been used to reveal the associations of hundreds
Bipolar disorder is a common psychiatric disorder with high heritability.Integrative gene expression data and genetic data would facilitate disease related gene dysfunction and related biological func
Both neurodegenerative and psychiatric disorder are common and severe CNS diseases with symptom similarity and comorbidity;however,the underlying biological processes across these disorders remain lar
Production of maternal haploids via intra-specific genotypes as the haploid inducer is routine and highly efficient in maize.However,the underlying mechanism of haploid induction (HI) is unclear.Triph
Parkinsons disease (PD) is a major neurodegenerative disease influenced by both genetic and environmental factors.Although previous studies have provided insights into the significant impact of geneti
Pancreatic beta cell dysfunction is a central role of developmemt of type 2 diabetes[1].And the chronic dislipidemia (lipotoxicity) and chronic hyperglycemia (glucotoxicity) have been postulated to co
Pathway enrichment algorithms are applied to detect biological pathways significantly related to a given set of genes associated with a specific disease or phenotype,experimental biology conditions.In
To explore the genetic mechanism of complex disease from a systems point of view,we developed different tools for analysis from variation,gene to pathway.As we all know,genome-wide association study (