基于辐射防护基因集的抗辐射损伤先导化合物预测

Prediction of lead compounds of radioprotectants based on radioprotective gene sets

目的利用基于网络的细胞反应印记整合文库(LINCS)计划的小分子化合物刺激不同细胞系的转录谱数据,构建算法评估化合物的抗辐射损伤活性,发现抗辐射损伤先导化合物。方法基于基因本体(gene ontology,GO)数据库构建辐射防护基因集,并基于基因集富集分析算法构建算法评估化合物的抗辐射损伤潜能,预测具有抗辐射损伤活性的先导化合物。结果建立了4个具有抗辐射损伤效果的辐射防护基因集,以及利用转录组数据预测抗辐射损伤先导化合物的方法;在32 560个小分子化合物中预测获得了299个抗辐射损伤先导化合物,其中17个已被报道具有抗辐射损伤活性。结论预测结果具有较好参考价值;预测方法速度快,成本低,可有效发现具有抗辐射损伤活性的化合物。

Objective To evaluate the radioprotective activity of compounds by proposing a new algorithm based on the transcriptome data from the Library of Integrated Network-based Cellular Signatures(LINCS)program,and to find lead compounds of radioprotective drugs.Methods Based on the gene ontology(GO)website,radioprotective gene sets were constructed and an algorithm for estimating the radioprotective potential of the compound was formulated based on the gene set enrichment analysis algorithm so as to predict the lead compounds of radioprotective drugs.Results Four radioprotective gene sets were established,and a new method for predicting radioprotective lead compounds using transcriptome data was established.Among the 32560 small molecule compounds,299 radioprotective lead compounds were found,17 of which had been reported to be resistant to radiation damage.Conclusion The prediction results are of good referential value,and the prediction method has the advantages of high speed,low cost and effective discovery of compounds with radioprotective activity.