流行病教學(xué)課件：分子流行病學(xué)

1、分子流行病學(xué)Molecular Epidemiology基本概念主要研究?jī)?nèi)容常用技術(shù)方法應(yīng)用實(shí)例提綱分子流行病學(xué)（Molecular Epidemiology）分子流行病學(xué)是闡明疾病和生物群體中醫(yī)學(xué)相關(guān)生物標(biāo)志的分布及其與疾病/健康的關(guān)系和影響因素，并研究防治疾病、促進(jìn)健康的策略與措施的科學(xué)。應(yīng)用先進(jìn)的技術(shù)測(cè)量生物標(biāo)志的分布情況，結(jié)合流行病學(xué)現(xiàn)場(chǎng)研究方法，從分子或基因水平闡明疾病的病因及其相關(guān)的致病過(guò)程。產(chǎn)生背景疾病防治面臨新的挑戰(zhàn)傳染性疾病病原體變異病原體耐藥性新發(fā)現(xiàn)傳染病慢性非傳染病暴露與發(fā)?。ㄋ劳觯╆P(guān)系的研究干預(yù)效果的研究人群易感性實(shí)驗(yàn)醫(yī)學(xué)病因假設(shè)分子生物學(xué)的發(fā)展理論突破DNA雙螺旋結(jié)構(gòu)

2、、遺傳中心法則、基因突變和基因表達(dá)調(diào)控等技術(shù)創(chuàng)新凝膠電泳技術(shù)、分子雜交、基因克隆以及微陣列芯片技術(shù)等吸煙肺癌傳統(tǒng)流行病學(xué)（宏觀、群體）分子流行病學(xué)（微觀、群體） ？分子流行病學(xué)與傳統(tǒng)流行病學(xué)的關(guān)系 傳統(tǒng)流行病學(xué)暴露疾病 ？ 前瞻性研究回顧性研究分子流行病學(xué)暴露疾病分子流行病學(xué)的特點(diǎn)與傳統(tǒng)流行病學(xué)不同研究的水平不同所解決的問(wèn)題不同與分子生物學(xué)的區(qū)別研究對(duì)象不同（人群）研究?jī)?nèi)容和方向不同（強(qiáng)調(diào)與宏觀暴露的關(guān)系）研究方法不同（流行病學(xué)思維和方法） 宏觀與微觀相結(jié)合！血清流行病學(xué)和人類(lèi)基因組流行病學(xué)血清流行病學(xué)（seroepidemiology）應(yīng)用血清學(xué)方法對(duì)血清中各種成分（包括抗原、抗體、生化物質(zhì)

3、等）的分布情況進(jìn)行研究，以闡明疾病及健康狀態(tài)在人群中的分布及其影響因素，并在采取預(yù)防措施后應(yīng)用血清學(xué)方法來(lái)考核其效果。人類(lèi)基因組流行病學(xué)（human genome epidemiology, HuGE）是應(yīng)用流行病學(xué)與基因組信息相結(jié)合的研究方法，開(kāi)展以人群為基礎(chǔ)的研究，評(píng)價(jià)基因組信息（基因或基因變異及其相應(yīng)編碼的產(chǎn)物）對(duì)人群健康和疾病的流行病學(xué)意義，是遺傳流行病學(xué)與分子流行學(xué)交叉的前沿學(xué)科?；谘逯猩飿?biāo)志和核酸生物標(biāo)志，二者可列入分子流行病學(xué)的范疇。分子流行病學(xué)是流行病學(xué)的一個(gè)分支(分子生物學(xué)？群體遺傳學(xué)？)分子流行病學(xué)的研究方法是隨著分子生物學(xué)的進(jìn)展不斷更新的吸煙肺癌傳統(tǒng)流行病學(xué)（宏觀、

4、群體）（微觀、群體） ？-omics的大數(shù)據(jù)（big data）時(shí)代系統(tǒng)流行病學(xué)生物標(biāo)志 (biological markers , biomarkers) 是指從暴露到疾病這個(gè)連續(xù)過(guò)程中可測(cè)量的，能反映功能或結(jié)構(gòu)變化的細(xì)胞、亞細(xì)胞、分子水平的物質(zhì)。分子生物標(biāo)志物 (molecular biomarkers)直接反映外來(lái)理化因素與細(xì)胞靶分子, 特別是生物大分子如核酸和蛋白質(zhì)的相互作用及其后果, 是生物標(biāo)志的主要部分。生物標(biāo)志 研究?jī)?nèi)容暴露測(cè)量 (E, ID, BED)效應(yīng)測(cè)量 (EBE, ASF, CD, PS)易感性測(cè)量 (DNA, RNA, Protein, Small Molecular

5、, Phenotype) 暴露標(biāo)志（exposure marker）（一）外暴露標(biāo)志主要指環(huán)境因素暴露，包括生物性因素和非生物性因素1、生物性病原因子的鑒定2、環(huán)境化學(xué)污染物（非生物因素）暴露（煙霧濃度）（二）內(nèi)暴露標(biāo)志1、傳染性疾?。?感染狀況抗原抗體檢測(cè)、病原體基因監(jiān)測(cè)2、慢性非傳染性疾病 進(jìn)入體內(nèi)的暴露劑量（可替寧濃度）（三）生物有效劑量標(biāo)志（加合物）效應(yīng)標(biāo)志（effect marker）效應(yīng)標(biāo)志（effect marker, MEF）：機(jī)體暴露后產(chǎn)生功能性或結(jié)構(gòu)性變化，并引起亞臨床階段和疾病發(fā)生過(guò)程的生物標(biāo)志，稱(chēng)為效應(yīng)標(biāo)志。早期生物效應(yīng)標(biāo)志：疾病前期生物標(biāo)志，細(xì)胞毒性反應(yīng)、染色體畸變、

6、DNA、RNA和蛋白表達(dá)改變結(jié)構(gòu)和功能改變 標(biāo)志：疾病發(fā)生標(biāo)志疾病臨床標(biāo)志：疾病發(fā)生后標(biāo)志，AFP、CEA、PSA、AST易感性標(biāo)志（susceptibility biomarker）（一）表型 DNA修復(fù)能力、端粒長(zhǎng)度（效應(yīng)、易感性都可）（二）表達(dá)譜 編碼基因、非編碼序列、蛋白表達(dá)（效應(yīng)、易感性都可）（三）小分子代謝物 脂類(lèi)、糖等的代謝產(chǎn)物（暴露、效應(yīng)、易感性都可）（四）基因組改變 多態(tài)性、copy數(shù)變異、表觀遺傳改變?cè)诒┞兑蛩刈饔孟?，宿主?duì)疾病發(fā)生、發(fā)展易感程度的生物標(biāo)志。spontaneous15%Gene-environment interaction78%pure genetic 5

7、%pure environment2%環(huán)境基因-+20%80%100%17%83%Schulte, 1994“Genetics loads gun, but Environments trigger”復(fù)雜疾病模式人類(lèi)基因組遺傳變異INS/DEL：插入/缺失 TRS：串聯(lián)重復(fù)序列(小衛(wèi)星和微衛(wèi)星)CNV：拷貝數(shù)變異INVERSION: 顛換SNP：?jiǎn)魏塑账岫鄳B(tài)性 生物學(xué)意義: 不同個(gè)體的基因99.9%是一樣的，但在序列上有極小(0.1%)的遺傳差異，其中主要是單核苷酸多態(tài)性(SNP)。這0.1%的差異授予你我他特有的表型；與個(gè)體對(duì)疾病的易感性、對(duì)藥物的反應(yīng)性差別有密切關(guān)系。 普遍性: SNP是指

8、在人群中出現(xiàn)的頻率1%的先天突變。SNP存在于整個(gè)人類(lèi)基因組中，每100bp以?xún)?nèi)就存在一個(gè)SNP，估計(jì)總數(shù)在數(shù)百萬(wàn)個(gè)。 對(duì)于人群研究而言的技術(shù)可及性和穩(wěn)定性: Platform and databaseWhy SNPs ?基因組遺傳變異與個(gè)體疾病易感性差異基因型檢測(cè)-危險(xiǎn)度評(píng)價(jià)、個(gè)體化治療分子流行病學(xué)常用技術(shù)及方法核酸技術(shù)（DNA & RNA）核酸體外擴(kuò)增 （PCR）核酸電泳圖譜 核酸酶切電泳圖譜 （RFLP）擴(kuò)增片斷長(zhǎng)度多態(tài)性（AFLP）核酸分子雜交 有原位雜交、斑點(diǎn)雜交、轉(zhuǎn)印雜交（包括Southern blot檢測(cè)DNA和Northern blot檢測(cè)RNA）等。核酸序列分析蛋白質(zhì)技術(shù) 蛋

9、白質(zhì)分離鑒定技術(shù)主要有：電泳、凝膠電泳、蛋白質(zhì)轉(zhuǎn)印雜交（Western blot）、色譜分析、蛋白質(zhì)測(cè)序等。酶學(xué)技術(shù) 包括定性和定量檢測(cè)，要求一定的條件。多位點(diǎn)酶電泳（multilocus enzyme electrophoresis, MEE）法。免疫學(xué)技術(shù)： 酶聯(lián)免疫吸附試驗(yàn)（ELISA）、熒光免疫試驗(yàn)（FIA）、放射免疫試驗(yàn)（RIA）、免疫細(xì)胞化學(xué)檢測(cè)（ICC）等；色譜（質(zhì)譜）技術(shù)： 高效液相色譜技術(shù)HPLC）、液相蛋白色譜技術(shù)（FPLC）等。個(gè)體化治療的途徑個(gè)體化治療患者避免-不必要的治療- 無(wú)效治療- 有害治療社會(huì)-資源有效利用- 減輕經(jīng)濟(jì)負(fù)擔(dān)預(yù)后相關(guān)- 治療是否合適預(yù)測(cè)相關(guān)- 治療

10、敏感性- 治療毒副反應(yīng)腫瘤特征-組織病理學(xué)類(lèi)型- 遺傳學(xué)改變- 基因表達(dá)- 蛋白表達(dá)病人特征-生殖系遺傳學(xué)改變(SNP)等Automated Liquid HandlerSpectroPREPTMNanoliter DispenserSpectroJETTMMALDI-TOF Mass SpectrometerSpectroREADERTMAutomated AnalysisSpectroTYPERTMMiniaturized BiochipSpectroCHIPTMMassARRAYIntegrated Components基因組研究的平臺(tái)中低通量檢測(cè)平臺(tái)PCR-RFLP、PCR-SSCP,

11、 PCR-based resequencing、TaqMantime-of-flight (MALDI) mass spectrometry、OpenArray基因組芯片SNP Chip（10K、20K、50K、100K、550K、1M、5M） 二代測(cè)序（NGS）Roche 454、Illumina Solexa、ABI SOLiD7900 Real-Time PCR TaqMan OpenArrayTaqMan OpenArray Genotyping PlatesHi-Seq iScanSQ 全基因組關(guān)聯(lián)研究(GWAS)通量高、樣本量大、多階段驗(yàn)證、質(zhì)控統(tǒng)計(jì)嚴(yán)格關(guān)聯(lián)結(jié)果真實(shí)可靠、重復(fù)性高！

12、Feero WG et al. New Eng J Med. 2010.I期II期III期SNPs數(shù)量篩選外部驗(yàn)證內(nèi)部驗(yàn)證二代測(cè)序（NGS）DNA（全基因組、外顯子組）Missing heritabilityUncovered variantsRare variationInDelCopy number variationMutationGermlineSomaticRNA（轉(zhuǎn)錄組RNA-Seq，sRNA）Epigenetic（Methyl-Seq，ChIP-Seq）特點(diǎn)：發(fā)現(xiàn)新的突變或其他結(jié)構(gòu)變異通量高、數(shù)據(jù)量大、質(zhì)控嚴(yán)格研究?jī)?nèi)容更多更廣基因組測(cè)序價(jià)格顯著下降Metagenomic Sequ

13、encingQin J et al. Nature. 2010 Karlsson FH et al. Nature. 2013 Faith JJ et al. Nature. 2013 Integrated GenomicsWhole Genome, Exome, Transcriptome, Small RNA, Methylation Nature 2013Nature 2013Nat Genet 2013分子流行病學(xué)在疾病預(yù)防和控制中的應(yīng)用（一）病因的探討及病因致病機(jī)制的研究 1、傳染病病因及流行規(guī)律的探討：研究已知或新發(fā)傳染病的病原體和流行特點(diǎn) 2、疫苗評(píng)價(jià)、非應(yīng)答者病原學(xué)診斷（基因診

14、斷）、基因分型研究分布特點(diǎn)，傳染源追蹤和傳播途徑確定觀察病原體在體內(nèi)持續(xù)時(shí)間和消長(zhǎng)規(guī)律案例：1990年7月，美國(guó)佛羅里達(dá)州一名AIDS婦女，懷疑在當(dāng)?shù)匾谎揽圃\所手術(shù)時(shí)感染了HIV，因?yàn)樵撛\所的牙科醫(yī)生也是一名AIDS患者。隨后當(dāng)?shù)匦l(wèi)生部門(mén)對(duì)該牙醫(yī)診治過(guò)的一千余名患者進(jìn)行HIV檢測(cè)，共發(fā)現(xiàn)7名HIV感染者有在該診所的就診史。傳染病傳播途徑分析Science,1992那么這7名HIV感染者是否與該牙醫(yī)有關(guān)呢？針對(duì)這一問(wèn)題可采取的分析方法：1、傳統(tǒng)流行病學(xué)分析方法：接觸者調(diào)查，分析感染者可能的危險(xiǎn)行為如輸血、靜脈吸毒、不潔性交史等。2、分子流行病學(xué)方法：比較HIV病毒基因序列的差異，分析不同病例感染

15、株遺傳關(guān)系的遠(yuǎn)近。Science,1992與牙醫(yī)感染株的遺傳變異性比較與對(duì)照感染株的遺傳變異性比較病人D當(dāng)?shù)貙?duì)照牙醫(yī)及病人A,B,C,E,G病人F進(jìn)化樹(shù)分析2.非傳染病的病因及病因致病機(jī)制的研究 例：血脂與心血管疾病 吸煙導(dǎo)致肺癌的作用機(jī)制 基因表達(dá)與疾病的發(fā)生與轉(zhuǎn)歸、個(gè)體化治療 Mitchell PS et al, PNAS. 2008 Chen X et al, Cell Res. 2008 研究背景： 2008年，中美科學(xué)家同步發(fā)現(xiàn)血漿中穩(wěn)定存在miRNAEsquela KA et al, Nat Rev Cancer. 2006Ryan BM et al, Nat Rev Cancer

16、. 2010Maria AC et al, Nat Rev Clin Oncol. 2011Kasinski AL et al, Nat Rev Cancer. 2011 Lujambio A et al, Nature. 2012 血漿miRNA的特點(diǎn):穩(wěn)定、非侵入性、可用于早期診斷和疾病監(jiān)測(cè)2萬(wàn)人4年隨訪(fǎng)早期診斷7名在基線(xiàn)調(diào)查抽血時(shí)正常但后期發(fā)生肺癌的個(gè)體比臨床平均提前了 14.8 個(gè)月隊(duì)列研究血漿miRNA表達(dá)譜作為肺癌早期診斷生物標(biāo)志物400肺癌病例/220對(duì)照10個(gè)血漿miRNAs靈敏度：0.93 特異度：0.90例：惡性腫瘤的形成過(guò)程及其影響因素前致癌物終致癌物DNA損傷基因突變癌

17、前病變惡性腫瘤活化解毒凋亡增生修復(fù)損傷遺傳易感性因素（二）疾病易感性的測(cè)定 （傳染病和非傳染?。㏕he Angelina effectMother died from BRCa at age 56, with a BRCA1 mutationShe carries the same BRCA1 mutation87% risk for BRCa and a 50% risk for ovarian cancerAt age 37 (2013), opted for mastectomyRaise public awareness for genomic testing and preventi

18、onIn march 2015, she had her ovaries and fallopian tubes removedHypothesis Driven Approach?。–andidate genes）Functional SNPs:5 flanking5UTRCoding3UTRTagging SNPs:Candidate genesCommon variantsLD valuesMetabolismCell cycleDNA repairApoptosisInflammationImmunity.Target pathwaySmoking Related DNA Damage

19、s and Repair PathwaysLinkage Disequilibrium (LD) based-mappingSequence-based selection approach based on SNP location and potentially functional relevanceGeneral Strategy for Selecting SNPsCategory of SNPsCoding SNPs (cSNPs): non-synonymous SNPs (nsSNPs) vs. synonymous SNPs (sSNPs)Non-coding SNPs (n

20、cSNPs): regulatory SNPs (rSNPs) vs. Other intronic SNPs (iSNPs)rSNPs: located at 3UTR/5UTR, promoter region or exon-intron boundaryNature, 2004Nature, 2004GWAS研究策略Stage IGenome-wide phase利用芯片對(duì)所有tagSNPs進(jìn)行初篩，找出差異位點(diǎn)Stage IIValidation phase I以10-7為水準(zhǔn)，在獨(dú)立大樣本人群中進(jìn)行驗(yàn)證Stage IIIValidation phase II對(duì)仍有意義的差異位點(diǎn)，進(jìn)

21、一步在外部人群驗(yàn)證Seven Years，More than 1600 published GWAS have reported about 2000 robust associations for 300 traitsNHGRI GWA CatalogBy 2013/6Reported GWAS of Lung CancerChromosome-log10 P value15q255p156p21Wang et al, Nat Genet, 2008Three loci indentified in lung cancer GWAS of European ancestryThorgeir

22、et al, Nature, 2008Hung et al, Nature, 2008Amos et al, Nat Genet, 2008McKay et al, Nat Genet, 2008Wang et al, Nat Genet, 2008Heterogeneity among ethnicitiesThe risk alleles of SNPs at 6p21 or 15q25 observed in European (CEU) were rare or not observed in Chinese (CHB), Japanese (JPT) or African (YRI)

23、 data in the HapMap database.LociSNPsBase changeCEUCHBJPTYRI6p21rs3131379 C T0.1020.0000.0000.0426p21rs3117582A C0.0800.0000.0000.04915q25rs8034191T C0.4330.0440.0110.14215q25rs16969968G A0.4250.0330.0110.00015q25rs1051730T C0.4000.0330.0110.110Only 5p15 locus is replicated in ChineseWu C et al. Can

24、cer Res. 2009Jin G et al. Carcinogenesis. 2009LociSNPsAllelesMAF*OR(95%CI) P 5p15rs2736100A C0.450/0.4181.16(1.03-1.30)0.0205p15rs402710C T0.293/0.3110.92(0.81-1.04)0.1616p21rs9295740G A0.188/0.1801.04(0.90-1.21)0.47315q25rs8034191T C0.028/0.0370.72 (0.501.02)0.06515q25rs16969968G A0.028/0.0350.84 (

25、0.591.21)0.34715q25rs1051730T C0.027/0.0330.86 (0.601.24)0.412*Minor allele frequency (case/control); Association results in Chinese population (1,221 cases and 1,344 controls )Similar results were reported in JapaneseMiki D et al. Nat Genet. 2010LociSNPsAllelesMAF*OR(95%CI) P 5p15rs2736100A C0.462/

26、0.3991.29(1.16-1.44)5.2E-066p21rs9295740G A0.163/0.1720.94(0.81-1.09)0.39715q25rs8034191T C0.028/0.0241.18 (0.84-1.65)0.33915q25rs1051730T C0.027/0.0231.20 (0.85-1.69)0.289*Minor allele frequency (case/control); Association results in Japanese population (1,004 cases and 1,900 controls )通過(guò)2331例肺癌病例和

27、3077例對(duì)照的初篩和后續(xù)多階段驗(yàn)證，發(fā)現(xiàn)9個(gè)與肺癌相關(guān)聯(lián)的風(fēng)險(xiǎn)區(qū)域，可以顯著提高肺癌危險(xiǎn)度預(yù)測(cè)效能，其中四個(gè)通過(guò)與吸煙的交互作用參與肺癌發(fā)生Hu Z et al. Nat Genet.2011Dong J et al. Nat Genet.2012鱗癌特異的位點(diǎn)多腫瘤共有危險(xiǎn)區(qū)域肺癌胃癌食管癌12q23.1 區(qū)域遺傳變異是肺鱗癌特異的遺傳易感區(qū)域從遺傳的角度區(qū)分肺鱗癌和腺癌Jin G, et al, Am J Hum Genet 2012 Dong J, et al, Plos Genet 2012 Significance of GWAS Findings (I)Partially elu

28、cidate susceptibility of Complex disease attributed to low penetrance variants. Risk factors : Lung cancer as a exampleCandidate gene approach: several loci such as CAPS8 D302N for breast cancer. GWAS approach: 200 loci and morePowerful approach!Significance of GWAS Findings (II)Provide new insights

29、 into the biology of complex disease. Science. 2012;337:1190-5About 90% GWAS identified variants are intronic (41.2%) or intergenic (52.5%).Hundreds of GWAS variants are related with long-distance regulation for distant gene targetsPrevious unknown pathogenic pathways: autophagy in Crohns disease, I

30、L-23 pathway (GWA studies: rewriting the story of IBD. Trends Genet. 2009). Clinical usage: GWAS findings have been or probably will be valuable in disease prediction (type 1 diabetes, AUC about 0.9), biomarker identification and disease subclassification (HNF1A-MODY) Significance of GWAS Findings (

31、IIIa)The analyses included 28 subjects with HNF1A-MODY, 294 with autoimmune diabetes, 187 with type 2 diabetes, and 24 with GCK-MODY and 198 nondiabetic control subjects.DIABETES CARE 2010Significance of GWAS Findings (IIIb)Clinical usage: GWAS findings have been or probably will be valuable in dise

32、ase prediction (type 1 diabetes, AUC about 0.9), biomarker identification and disease subclassification (HNF1A-MODY) , treatment selection (ITPA gene deficient)ribavirin induced haemolytic anaemia:利巴韋林引起溶血性貧血ITPA deficiency protects against clinically significant decline inHb concentration induced b

33、y HCV anti-viral treatment.Nature 464, 405408 (2010)Risk of myopathy in chronic simvastatin use :他汀類(lèi)藥物導(dǎo)致的肌病 N. Engl. J. Med. 359, 789799 (2008)Clinical usage: GWAS findings have been or probably will be valuable in disease prediction (type 1 diabetes, AUC about 0.9), biomarker identification and dis

34、ease subclassification (HNF1A-MODY) , treatment selection (ITPA gene deficient), and drug dosing (Statin-Induced Myopathy). Significance of GWAS Findings (IIIc)rs4149056, 影響代謝 FindingsChallengesStrategiesMore loci / rare variantClarify the biological mechanism of GWAS variants in cancer development

35、and determine the causal variants/ genesTranslate these genetic loci from bench to bedside Discovering Missing HeritabilityInternational collaboration: Pooling GWAS data via consortium and sharing pipelines and samplesCopy number variations (CNVs): calling based on GWAS chips, but hard to clean the

36、dataEpistasis and gene-environment interaction: a new wave of GWAS based on carefully designed large representative nested case-control study / stable internal exposure instead of external exposureWhole genome sequencing based GWAS and rare variants: a ideal but cost-dependent approachDetermining Ca

37、usal Variants/GenesGenetic refinement Target sequencing Fine-mapping studyGWASFunction assessment In vitro In vivoExpression analysis eQTLs Proteinomics ENCODE functional annotationMarker SNPsCausalSNPsGWAS109 SNPs (P 10-5)Knock-down/Knock-out果蠅、斑馬魚(yú)等模式動(dòng)物同源基因表型分析相應(yīng)組織表達(dá)譜分析或eQTL分析SNPseQTL-mappingKO mod

38、elGenesSpermatogonium blockspermiogenesisanomalyWTspermiogenesisanomalyPost-GWAS易感基因表型研究流程McCarthy et al, Nat Rev Genet. 2008Identification of susceptibility lociNovel biological insightsImproved measures of individual aetiological processTherapeutictargetsBiomarkersPreventionDiagnosticsPrognosticsT

39、herapeutic optimizationTranslating From Bench to Bedside Calories from Soft Drinks Do They Matter?A Trial of Sugar-free or Sugar-SweetenedBeverages and Body Weight in ChildrenN Engl J Med. 2012 Oct 11;367:1397-406A Randomized Trial of Sugar-Sweetened Beverages and Adolescent Body WeightN Engl J Med.

40、 2012 Oct 11;367:1407-16Nurses Health Study (NHS, 6934 women)Health Professionals Follow-up Study (HPFS, 4423 men)Beverage Intake Genetic-predisposition score from 32 BMI-associated loci Womens Genome Health Study (WGHS, 21740 women)Replication stage:Initial stage:BMIObesityN Engl J Med. 2012 Oct 11

41、; 367(15):1387-96.Sugar-Sweetened Beverages and Genetic Risk of ObestiyDifference in BMI Associated with One Serving of a Sugar-Sweetened Beverage per DayN Engl J Med. 2012 Oct 11;367(15):1387-96.Translating From Bench to Bedside For complex disease, Art and Skill in the whole process！- 沒(méi)有“解決”方案！(3

42、vs. 17 years)Do you really understand the power and utility of “family history”, maybe the only useful information from patients answerHowever, in a prediction model, family history usually add little AUC useless ? The balance of noise and useful information for complex diseaseClinical utility: targ

43、eted PSA screenSun et al, 2013Genetic score and family historyLimitations of family historyIndirect measurement based on relatives, not purely geneticDifficult to be measured accurately, affected by many factorsDefinitions (first-degree or second-degree)Family sizeCommunication among relativesLess i

44、nformative, either yes or no (vast majority)Some flaws of family historyBrothers have the same family history risk although they only share 50% of the same genes on averageFamily history is uninformative in populations with historically low incidenceGenetic score has a better precisionSun et al, 2013Genetic score could be different for brothers二代測(cè)序揭示腫瘤突變譜The landscape of mutational signaturesAlexandrov et al, Nature 20137,042 cancers; 4,938,362 mutations; 20 mutational