发表机构
University of Ibadan; Worcester Polytechnic Institute（伊巴丹大学; 伍斯特理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出网络结构贝叶斯层次模型，结合GMRF与马蹄形先验及共轭吉布斯采样，在癌症药物基因组学中识别稀疏、可解释且外部支持的基因-药物敏感性关联，并支持组织特异性分析。
AI 中文摘要
我们开发了一种网络结构贝叶斯层次模型，用于基因组改变与定量治疗反应表型之间的稀疏关联映射。该框架结合了高斯马尔可夫随机场先验（该先验在通路连接的基因间借用信息）、诱导稀疏性的全局-局部马蹄形先验，以及不需要Metropolis-Hastings步骤的共轭吉布斯采样器。尽管该方法广泛适用于具有已知预测器网络的高维设置，我们使用癌细胞系药物敏感性数据对其进行了验证。应用于GDSC2（N=951个细胞系，G=219个驱动基因，D=295种药物），该模型识别出126个基因-药物关联（占64,605对中的0.195%），集中在EZH2（45种药物，均为敏感性方向，平均效应-0.911 lnIC50）和KMT2D（36种药物，均为敏感性方向，平均效应-0.496 lnIC50）。这些标记在独立的PRISM筛选（1,518种化合物）中获得外部支持，其中KMT2D实现完全方向复制（36/36），EZH2部分复制（8/12）。五折交叉验证的预测对数似然确认了每个先验层的价值：完整模型每折比无网络消融模型高出+3,109对数单位，比无马蹄形消融模型高出+14,039对数单位，且在各折中保持一致。三种情景下的模拟显示，完整模型在整个过程中实现最高精度和最低错误发现率，而网络先验在网络结构信号下提高了敏感性恢复。组织分层扩展识别出连贯的亚组细化，包括肺特异性EGFR抑制剂敏感性和皮肤特异性BRAF-Dabrafenib敏感性。这些结果表明，该框架能够识别稀疏、可解释、外部支持的药物敏感性标记，同时能够对共享效应的组织特异性偏离进行原则性研究。
英文摘要
We develop a network-structured Bayesian hierarchical model for sparse association mapping between genomic alterations and quantitative treatment-response phenotypes. The framework combines a Gaussian Markov random field prior that borrows strength across pathway-connected genes, a global-local horseshoe prior inducing sparsity, and a conjugate Gibbs sampler requiring no Metropolis-Hastings steps. Though broadly applicable to high-dimensional settings with known predictor networks, we validate it using cancer cell-line drug-sensitivity data. Applied to GDSC2 ($N=951$ cell lines, $G=219$ driver genes, $D=295$ drugs), the model identifies 126 gene-drug associations (0.195\% of 64{,}605 pairs), concentrated in EZH2 (45 drugs, all sensitivity-direction, mean effect $-0.911$ $\ln$IC50) and KMT2D (36 drugs, all sensitivity-direction, mean effect $-0.496$ $\ln$IC50). These markers show external support in an independent PRISM screen (1{,}518 compounds), with KMT2D achieving complete directional replication (36/36) and EZH2 partial replication (8/12). Five-fold cross-validated predictive log-likelihood confirms each prior layer's value: the full model outperforms the no-network ablation by $+3{,}109$ log-units per fold and the no-horseshoe ablation by $+14{,}039$ log-units, consistently across folds. Simulations under three scenarios show the full model achieves the highest precision and lowest false-discovery rate throughout, while the network prior improves sensitivity recovery under network-structured signal. A tissue-stratified extension identifies coherent subgroup refinements, including lung-specific EGFR-inhibitor sensitivity and skin-specific BRAF-Dabrafenib sensitivity. These results show the framework identifies sparse, interpretable, externally supported drug-sensitivity markers while enabling principled investigation of tissue-specific departures from shared effects.
CommentsThe manuscript is under review with Biostatistics (Oxford Academic)
