基于大量ChIP数据集的果蝇顺式调控模块的从头预测

De novo prediction of cis-regulatory modules in D. melanogaster based on a large number of ChIP datasets

通过整合果蝇已有的ChIP数据集,采用模体发现算法FisherNet及高性能并行的模体聚类算法CLIMP对果蝇的顺式调控模块进行从头预测.与已知的顺式调控模块进行比对分析,结果表明该方法预测结果覆盖了数据集中已知顺式调控模块的82.93%,证明该方法具有一定的普适性.与较新的DePCRM算法进行比较,结果表明本算法在从头预测顺式调控模块上速度更快、精度更高.

The ChIP datasets of D. melanogaster are integrated and used to predict the cis-regulatory modules by employing the motif discovery algorithm Fisher Net and the motif clustering algorithm CLIMP. By comparing with the known cis-regulatory modules,the prediction results recover 82.93% of the known cis-regulatory modules,which shows that the method has a certain universality. Comparing with the DePCRM algorithm,it is found that this algorithm is faster and more accurate than DePCRM for predicting the cis-regulatory modules.