|||
基因组坏了,修还是不修?这是个问题。
顺铂(cisplatin)是常用化疗药,会进入细胞核与DNA结合,导致基因组损伤,进而引起细胞死亡。DNA修复通路是维持基因组稳定性的重要机制,但这套机制对顺铂引起的基因组损伤也可以进行修复,减少因基因组破坏导致的癌细胞死亡;同时顺铂引起的基因组变化被修复成了新的“突变”,增加了癌细胞进化出抗药性的可能性。我们的分析发现,在膀胱癌(BLCA)中,DDB1(损伤特异的DNA结合蛋白)的高表达预示着顺铂药物无效,从整体生存期来看DDB1的高表达也也预示着较差的预后。可以说DNA修复通路是癌症发生发展的双刃剑,DNA修复基因与顺铂药物响应的复杂关系有待深入研究。
==============================
判断化疗药物临床响应的好坏是癌症治疗的重要问题,从药物成本和总体疗效来看,化疗药仍是一线治疗方案的首选。但并不是所有病人对化疗的响应一致。除了传统的临床检查指标外,从分子层次建立化疗药物临床响应的预测模型、阐明抗药机制对癌症的治疗具有重要意义。我们系统整理了TCGA的药物临床响应和多模态组学数据,建立了一套严格的机器学习分析框架对基于组学数据的预测模型的性能进行了系统的评价,结果表明:
1)截止2015年年底数据,同一种药物的药物响应与组学数据都比较完整的记录仍不多,最多的几种化疗药cisplatin、5-fu等也仅有100多个样本,在单个癌症类型中样本数仅有数十个;另外,“无响应”与“响应”组的样本数很不平衡,有数倍的差异。数据集的这些特点给构建预测模型带来了很大的难度;
2)针对数据的特点,我们设计了基于系数回归(elastic net)和重采样(bootstrapping)的计算评价框架,尽可能避免过学习,相对客观的评价预测的性能;
3)从预测结果来看,总体预测性能并不理想,仅有膀胱癌-顺铂-基因表达、乳腺癌-紫杉醇-miRNA表达等几个数据集可以达到较好的预测性能;
4)利用elastic net进行特征选择,可以选出若干与药物临床响应相关的重要基因,比如DDB1(DNA repair pathway)高表达预示着响应差,从生存分析的结果来看,DDB1高表达与预后差相关,顺铂对基因组具有破坏性,DDB1基础表达水平高可能通过削弱顺柏的基因组破坏作用进而产生药物抗性;另外还发现DLL4、INST5、HNRNPA3-HNRNPA3P1等很有意思的特征基因;
5)借鉴pan-cancer分析的思路,我们也对单种药物也进行了跨癌症类型的分析,对预测性能会有影响,但未能得出有规律的变化。
Evaluating the molecule based prediction of clincial drug responses in cancer [LINK]
Zijian Ding, Songpeng Zu, Jin Gu
Dataset: http://bioinfo.au.tsinghua.edu.cn/member/jgu/drug_response
Molecule-based prediction of drug response is one major task of precision oncology. Recently, large-scale cancer genomic studies, such as The Cancer Genome Atlas (TCGA), provide the opportunity to evaluate the predictive utility of molecular data for clinical drug responses in multiple cancer types. Here, we firstly curated the drug treatment information from TCGA. Four chemotherapeutic drugs had more than 180 clinical response records. Then, we developed a computational framework to evaluate the molecule based predictions of clinical responses of the four drugs and identify the corresponding molecular signatures. Results show that mRNA or miRNA expressions can predict drug responses significantly better than random classifiers in specific cancer types. A few signature genes are involved in drug response related pathways, such as DDB1 in DNA repair pathway and DLL4 in Notch signaling pathway. Finally, we applied the framework to predict responses across multiple cancer types and found that the prediction performances get improved for cisplatin based on miRNA expressions. Integrative analysis of clinical drug response data and molecular data offers opportunities for discovering predictive markers in cancer. This study provides a starting point to objectively evaluate the molecule-based predictions of clinical drug responses.
=========================
附:投稿过程
这次可以说是最近第一次写非方法类的研究论文,主要研究用组学数据预测化疗药物临床响应。
从投稿过程来看,没有新的组学数据或者功能性验证实验,计算发现很难得到基础医学类杂志的杂志,几个好一点的杂志都没有送审。
第一个期刊是生物综合类杂志核酸研究NAR,编辑部未送审,主要理由是缺乏general interests,2016/01/28
While your study is very interesting, the editors believe that this paper is most appropriate for a journal more specifically focused on cancer therapeutics and drug validation. Therefore, we are returning this decision to you so that you may proceed with submission to a different journal. This decision does not reflect our assessment of the quality of this work but rather the belief that it is more appropriate for a different publication venue.
第二个期刊是癌症类专业期刊Cancer Research,1月28日投稿,因为格式问题来回了几次,2月4日进入Under Review(应该是在编辑手里),2月16日中午Decision Pending,稍后收到编辑部未送审通知。
第三个期刊是Nature Communications,2月17日投稿,2月29日收到编辑部未送审通知。
第四个期刊投Bioinformatics,格式改成了Discovery Notes,3月1日投出,3月6日通过编辑部进入Awaiting Reviewer Assignment,3月14日Under Review,3月28日Awaiting Decision;4月3日Major Decision,4月28日修回,5月4日编辑部审查后进入Under Review,5月23日Awaiting Decision,5月27日Accepted
Archiver|手机版|科学网 ( 京ICP备07017567号-12 )
GMT+8, 2024-11-23 09:34
Powered by ScienceNet.cn
Copyright © 2007- 中国科学报社