CSI-ANN

[…] marked by histone modifications were identified and annotated to refseq genes by diffreps () with default settings., enhancers were defined as h3k4me3low + h3k27achigh and predicted using csi-ann () with normalized h3k4me3 and h3k27ac signals as inputs. the predicted enhancers and h3k27ac signal were then used as inputs to predict super-enhancers using rose […]

[…] marked by histone modifications were identified and annotated to refseq genes by diffreps () with default settings., enhancers were defined as h3k4me3low + h3k27achigh and predicted using csi-ann () with normalized h3k4me3 and h3k27ac signals as inputs. the predicted enhancers and h3k27ac signal were then used as inputs to predict super-enhancers using rose […]

[…] prediction probability indicating how likely this genetic variant is a risk variant in a given disease., enhancers were predicted using the chromatin signature inference by artificial neural network csi-ann algorithm. the input to the algorithm is the normalized chip-seq signals of three histone marks (h3k4me1, h3k4me3, and h3k27ac). the algorithm combines signals of all histone marks and uses […]

[…] [], which annotated human genome characteristics for eight cell lines and reported higher recognition performance compared with [] as indicated by the area under curve (auc)., in particular, csi-ann [] is one of the first enhancer classification systems that rely on an ann using chromatin signatures as input. putative enhancers derived from human cd4t cell data from wang et al. [] based […]

[…] zero signals in both data sets were excluded for computing pearson correlation coefficient., we used “h3k4me3lo + h3k27achi” to define enhancers. specifically, enhancers were predicted using the csi-ann algorithm and normalized h3k4me3 and h3k27ac mowchip-seq signals across the genome. the coordinates of the predicted enhancers and h3k27ac mowchip-seq data were then used as the input […]

[…] see supplementary data for the description of additional data sources., we recently developed a computational method, termed chromatin signature identification by artificial neural network (csi-ann), for identifying functional dna elements using their chromatin signatures (). the framework consists of a data transformation and a feature extraction step using fisher discriminant […]