药物预测oncoPredict与pRRophetic

pRRophetic是一个用于预测肿瘤药物敏感性的R包。它的算法是基于基线基因表达水平和体外细胞系对药物的敏感性之间的关系建立的模型。它使用ridge回归方法来预测临床样本的药物反应。而oncoPredict是pRRophetic的一个升级版本，它可以从细胞系筛选数据中预测体内或癌症患者的药物反应和生物标志物。

尽管两种方法的目标都是进行肿瘤药物敏感性预测，但它们之间存在一些差异。oncoPredict是pRRophetic的升级版，它可能提供更准确的预测结果。此外，oncoPredict还可以预测药物反应和生物标志物，为研究人员提供更全面的信息。

pRRophetic/pRRophetic at master · paulgeeleher/pRRophetic (github.com)

himawork/pRRophetic (gitee.com)

oncoPredictoncoPredict: an R package for predicting in vivo or cancer patient drug response and biomarkers from cell line screening data - PMC (nih.gov)

Example use case 

To assess the model performance of the calcPhenotype function, we imputed drug response in an ovarian cancer clinical dataset, which had both patient response to paclitaxel and gene expression microarray data (GSE51373) . We used the CTRP dataset to train a model of paclitaxel drug response. We then stratified patients by their observed clinical outcome (defined by the trial as resistant or sensitive to paclitaxel treatment) and compared the predicted sensitivity scores (Figure 3). We see that imputed paclitaxel sensitivity is able to correctly stratify patients into their responder/non-responder categories (P = 0.0075 by t-test). Additionally, the area under the receiver operating characteristic (ROC) curve was 0.79, which is consistent with the performance of the models of the original paper

Please check https://github.com/maese005/oncoPredict and OSF | oncoPredict for data

pRRophetic（太老了，不如用oncoPredict）

pRRophetic包并不使用我们介绍的Cancer Therapeutics Response Portal (CTRP) 和 Genomics of Drug Sensitivity in Cancer (GDSC) 两个数据库资源，而是使用 Cancer Genome Project (CGP)计划里面的表达量矩阵和药物处理信息，该数据库有138 anticancer drugs against 727 cell lines。

Expression pattern of secretory‐cell‐related transcriptional signatures in colon adenocarcinomas defines tumor microenvironment characteristics and correlates with clinical outcomes - PMC (nih.gov)https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9980301/该文献联合了the Cancer Therapeutics Response Portal (CTRP) database 数据库数据使用

R包pRRophetic运行时出错：Error in if (class(testExprData) != "matrix") stop("ERROR: \"testExprData\" must be a matrix.") : the condition has length > 1

解决：

方法1：R版本过新，退回到4.1.3以前。

方法2：基因表达量预测药物反应的R包pRRophetic近期报错解决方案 · Issue #2691 · ixxmu/mp_duty (github.com)