搜索

Genomic Signature of Homologous Recombination Deficiency in Breast and Ovarian Cancers
乳腺癌和卵巢癌中的同源重组缺陷基因的特征   

下载 PDF 引用 收藏 提问与回复 分享您的反馈 Cited by

本文章节

Abstract

Homologous recombination deficiency, mainly resulted from BRCA1 or BRCA2 inactivation (so called BRCAness), is found in breast and ovarian cancers. Detection of actual inactivation of BRCA1/2 in a tumor is important for patients’ treatment and follow-up as it may help predicting response to DNA damaging agents and give indication Homologous recombination deficiency, mainly resulted from BRCA1 or BRCA2 inactivation (so called BRCAness), is found in breast and ovarian cancers. Detection of actual inactivation of BRCA1/2 in a tumor is important for pat for genetic testing. This protocol describes how to detect impairment of homologous recombination based on the tumor genomic profile measured by SNP-array. The proposed signature of BRCAness is related to the number of large-scale chromosomal breaks in a tumor genome calculated after filtering and smoothing small-scale alterations. The procedure strongly relies on good quality SNP-arrays preprocessed to absolute copy number and allelic content (allele-specific copy number) profiles. This genomic signature of homologous recombination deficiency was shown to be highly reliable in predicting BRCA1/2 inactivation in triple-negative breast carcinoma (97% accuracy; for more details, see Popova et al., 2012) and predictive of survival in ovarian carcinoma (unpublished data). Authors are grateful to Dominique Stoppa-Lyonnet, Anne Vincent-Salomon, Thierry Dubois, and Xavier Sastre-Garau for their contributions. (Patent was deposited: Reference number EP12305648.3, June 7, 2012)

Data and Software

I.   Data:

  1. Whole genome SNP-array profile of a tumor. Affymetrix and Illumina are the major platforms providing high quality SNP-array chips and software for primary normalization. Specific protocols are available in manufacturers' websites www.affymetrix.com or www.illumina.com. SNP array profiles have to be further processed to absolute copy number and allelic content profiles by some software for mining SNP array profiles. Good examples of properly processed SNP-array profiles are in Cancer Cell line collection of Sanger Institute.
    http://www.sanger.ac.uk/cgi-bin/genetics/CGP/cghviewer/CghHome.cgi
  2. SNP-array like profiles obtained from Next Generation Sequencing (NGS) also could be used in this protocol if they are processed to absolute copy number and allelic content profiles; however, we do not consider it here in details.


II.  Software:

  1. Software for primary normalization of SNP-arrays: Genotyping Console, ChAS (both www.affymetrix.com), Genome Studio (www.illumina.com), Aroma package (www.aroma-project.org), tQN for quantile normalization of Illumina arrays (Staaf et al., 2008).
  2. Software for mining SNP array profiles to obtain absolute copy numbers and allelic contents (allele-specific copy numbers), such as GAP (Popova et al., 2009), PICNIC (Greenman et al., 2010), ASCAT (Van Loo et al., 2010), GPHMM (Li et al., 2011), TAPS (Rasmussen et al., 2011), Absolut (Carter et al., 2012), etc.
  3. The GAP method and further data processing were realized in R environment (www.r-project.org). However, any other language could be used to perform this analysis, including MatLab, Java, C++, etc.

Equipment

  1. Computers (2 GHz, 2 G RAM, Intel Core 2, 40 G HD)

Procedure

I.   Preprocessing of SNP array data

  1. Normalize .CEL files by the appropriate software depending on the array platform.
  2. Export the normalized data: chromosome, position, Log_R_Ratio (Illumina) or Log_2_Ratio (Affymetrix), B allele frequency (BAF, Illumina) or Allelic Difference (AD, Affymetrix) into a text file.
  3. Process SNP-arrays by (for example) GAP method (Popova et al., 2009) to obtain (Birkbak et al., 2011) estimation of normal contamination and (Carter et al., 2012) absolute copy number and allelic content profiles (Table 1).
    Table 1. Segmented tumor genomic profile (a fragment from Affymetrix OncoScan 300K)
    Position Start
    Position End
    Chromosome*
    Length SNPs
    Copy Number
    Major Allele
    59369
    115065314
    1
    13869
    2
    1
    115067829
    121049277
    1
    639
    3
    2
    143701096
    144299541
    1.5
    47
    3
    2
    144337336
    144989346
    1.5
    19
    0
    0
    145008423
    148551158
    1.5
    252
    3
    2
    148565769
    152700090
    1.5
    554
    6
    5
    * 1 stands for p arm and 1.5 stands for q arm of chromosome 1; pericentric region is indicated in red.


II.  Quality control

  1. Quality control of measured SNP-array profile: software for primary data normalization usually provides quality index for each chip; chips indicated to have marginal quality have to be excluded from further analysis; indication of quality cut-offs could be found in corresponding User Guides.
  2. Contamination of tumor sample by normal stromal cells: sample with more than 65% of predicted normal cells admixture have to be excluded from further analysis (Note: Measured tumor sample usually represents a mixture of tumor and normal cells in different proportions, which results in different contrast in the measured SNP-array profile; 60-70% of normal contamination is at the limit of current recognition techniques).
  3. Quality control of copy number and allelic content recognition: pattern of copy number alterations (CNAs) in a tumor genome have to be "interpretable", meaning, copy number variation and allelic imbalance profiles have to be consistent. Unfortunately, there is no reliable measure of such consistency developed; we used manual control of recognition based on the GAP plots (Figure 1). GAP plot of a tumor genome is a two dimensional representation of segmented SNP array profiles, where each circle represents a segment (Popova et al., 2009). Clear and regular structure of the GAP plot indicates consistency (Figure 1A), while chaotic structure indicates inconsistency (Figure 1B). Samples with inconsistent profiles have to be excluded from further analysis.


    Figure 1. GAP plots for two tumor samples measured by Affymetrix OncoScan 300K representing (A) high quality and (B) low quality profiles. GAP plot of a tumor genome is a two dimensional representation of segmented SNP array profiles, where each circle represents a segment (Popova et al., 2009). Tumor samples are from GEO database (GSE28330, Birkbak et al., 2012).

  4. Quality control of adequate detection of chromosomal breaks: Highly contaminated tumor samples together with unspecific variation in SNP array profiles often result in false positive chromosomal breaks detected by segmentation algorithms; the sample need to be discarded in the case of large number of false positive breaks. Adequate formal procedure for this type of quality control is not yet developed. We performed rough visual estimation of consistency of detected breaks in copy number variation and in recognized copy number profiles.
    Note: Poor quality sample comprises around 10-15% of hybridized samples, including low tumor content, poor hybridization, low recognition quality, etc.


III. Calculating the number of large-scale chromosomal breaks from segmented profile (Table 1, Figure 2):

Note: Here we describe how to estimate the number of chromosomal breaks related to homologous recombination deficiency; filtering of variation is performed only for the purpose of estimation of breakpoints number and has no relation to particular alterations whatever important they are.

  1. Filtering out micro-variation: The size of micro-variation S_micro is the lower limit of the detectable somatic alteration size, which is dependent on the SNP density in the array; for example, we used 50 SNPs for Affymetrix SNP 6.0; 30 SNPs for Illumina 600K; etc.
    Note: Main reason for this filtering is that micro-variations are often linked to germline copy number variations.
    1. Exclude from the segmented genomic profile all segments less than S_micro SNPs and link adjacent segments if they have identical Copy Numbers and Major Alleles.
  2. Filtering out and smoothing small-scale variation: The size of small-scale variation, S_small < 3 Mb, was defined in (Popova et al., 2012).
    Note: Main reason for this definition is that starting from 3 Mb chromosomal breaks follow a Poisson distribution, i.e. are independent from each other; while the small-scale segments tend to cluster in discrete chromosomal regions.
    1. Order small-scale segments according to the size.
    2. Exclude from the segmented genomic profile the smallest segment and link adjacent segments if they have identical Copy Numbers and Major Alleles.
    3. Repeat filtering and smoothing until the last small segment.
      Note: The way of filtering and smoothing small-scale variations has a minor effect on the resulting profile.
  3. Calculating number of Large-scale State Transitions (LSTs) of the size S Mb: LST_SMb is defined as a chromosomal break (change in copy number or allelic content) between two adjacent (< 3 Mb in between) segments >= S Mb each; number of LSTs is calculated directly from the segmented genomic profile after filtering and smoothing of small-scale variation (Figure 2).
    1. Annotate chromosomal breaks as follows: If two segments from the same chromosome arm differ in Copy Number or in Major Allele, and are >= SMb in size, and the distance between the segments is < 3 Mb, the break is annotated as LST_SMb;
    2. Calculate number of LST_SMb (S = 3, 4,…, 11 Mb) in a tumor genome.
      Note: Centromeric breaks are not taken into account.


      Figure 2. Example of genomic profile of one chromosome with detected copy numbers and LSTs. LRR: log R ratio profile; BAF: B allele frequency profile; GT: segmental genotypes recognized by GAP; LST_10 Mb: black arrows point to LST_10 Mb detected. The black under-line shows large-scale segments obtained after filtering and smoothing small-scale variations seen in the GT profile. Chromosome 3 of a tumor sample from GEO database is shown (GSE28330, Birkbak et al., 2012).


IV. Estimation of tumor ploidy:

  1. Estimate DNA index for a tumor genome as an average copy number in a genome divided by 2.
  2. Estimate chromosome counts in a tumor genome as a sum of copy numbers at pericentric regions of each chromosome arm (Table 1), following the rules:
    1. If the size of a segment in pericentric region is >= 1.5 Mb (or 500 SNPs for Affymetrix SNP6.0), the number of copies of corresponding chromosome arm is set to that of the segment;
    2. If the size of a segment in pericentric region is < 1.5 Mb, chromosome arm count is replaced by its average copy number.
      Note: Chromosome number is estimated after filtering micro-variation.
  3. Estimate tumor ploidy following the rule: tumor ploidy is estimated to be 2 (near-diploid genome) if DNA index < 1.3 and chromosome counts < 60; tumor ploidy is estimated to be 4 (near-tetraploid genome) if DNA index >= 1.3 and chromosome counts >= 60.
    Note: This attribution is obtained for breast and ovarian cancer genomes based on the analysis of a large number of tumor genomes, (Popova et al., 2012). Genomes with ambiguous attribution of ploidy represented less than 5% of all cases considered. Other cancers might have different genomic evolution and the thresholds for ploidy attribution might need to be adjusted.


V.  Signature of homologous recombination deficiency in a tumor genome

Based on the analysis of a large series of breast cancers with known status of BRCA1/2 genes the number of LST_6,7,8,9,10 Mb were found to represent effective discriminating features with naturally defined ploidy-specific cutoffs, which allowed prediction of BRCA1/2 inactivation with high accuracy and precision (Table 2). Testing the signature on ovarian cancer showed LST_6,7 Mb to be the most efficient prediction features with similar to breast cancer cohort cut-offs.

  1. Tumor genome is annotated as homologous recombination deficient if number of LSTs in a tumor genome is higher than corresponding ploidy-specific cut-off (Table 2).
    Note: Tumors with borderline LST number could be false positives due to false positive breaks detected in the genome. Inconsistency among LST_6, 7, 8, 9, 10 Mb predictions are rare.
    Table 2. Cut-offs for LST number predicting BRCAness in breast cancer
    LST_S Mb, S
    Ploidy 2: (p=68, N=182)
    Ploidy 4: (P=53, N=123)
    Cut-Off*
    FPR
    TPR
    Cut-Off
    FPR
    TPR
    6
    19 (17)
    0.04
    0.99
    32 (32)
    0.10
    1
    7
    17 (15) 0.05
    0.99
    29 (27)
    0.07
    0.98
    8
    14 (14)
    0.06
    1
    26 (26)
    0.08
    1
    9
    14 (11)
    0.04
    0.99
    25 (19)
    0.07
    0.98
    10
    11 (11)
    0.07
    1
    22 (18)
    0.06
    0.98
    *Cut-offs correspond to max (TPR-FPR); cut-offs in parenthesis correspond to 100 sensitivity.
    P: Numbers of BRCA1/2 mutated tumors; N: Number of BRCA1/2 wild type or not tested tumors;
    TPR: True positive rate; FPR: False positive rate.

References

  1. Birkbak, N. J., Wang, Z. C., Kim, J. Y., Eklund, A. C., Li, Q., Tian, R., Bowman-Colin, C., Li, Y., Greene-Colozzi, A., Iglehart, J. D., Tung, N., Ryan, P. D., Garber, J. E., Silver, D. P., Szallasi, Z. and Richardson, A. L. (2012). Telomeric allelic imbalance indicates defective DNA repair and sensitivity to DNA-damaging agents. Cancer Discov 2(4): 366-375.
  2. Carter, S. L., Cibulskis, K., Helman, E., McKenna, A., Shen, H., Zack, T., Laird, P. W., Onofrio, R. C., Winckler, W., Weir, B. A., Beroukhim, R., Pellman, D., Levine, D. A., Lander, E. S., Meyerson, M. and Getz, G. (2012). Absolute quantification of somatic DNA alterations in human cancer. Nat Biotechnol 30(5): 413-421.
  3. Greenman, C. D., Bignell, G., Butler, A., Edkins, S., Hinton, J., Beare, D., Swamy, S., Santarius, T., Chen, L., Widaa, S., Futreal, P. A. and Stratton, M. R. (2010). PICNIC: an algorithm to predict absolute allelic copy number variation with microarray cancer data. Biostatistics 11(1): 164-175.
  4. Li, A., Liu, Z., Lezon-Geyda, K., Sarkar, S., Lannin, D., Schulz, V., Krop, I., Winer, E., Harris, L. and Tuck, D. (2011). GPHMM: an integrated hidden Markov model for identification of copy number alteration and loss of heterozygosity in complex tumor samples using whole genome SNP arrays. Nucleic Acids Res 39(12): 4928-4941.
  5. Popova, T., Manie, E., Rieunier, G., Caux-Moncoutier, V., Tirapo, C., Dubois, T., Delattre, O., Sigal-Zafrani, B., Bollet, M., Longy, M., Houdayer, C., Sastre-Garau, X., Vincent-Salomon, A., Stoppa-Lyonnet, D. and Stern, M. H. (2012). Ploidy and large-scale genomic instability consistently identify basal-like breast carcinomas with BRCA1/2 inactivation. Cancer Res 72(21): 5454-5462.
  6. Popova, T., Manie, E., Stoppa-Lyonnet, D., Rigaill, G., Barillot, E. and Stern, M. H. (2009). Genome Alteration Print (GAP): a tool to visualize and mine complex cancer genomic profiles obtained by SNP arrays. Genome Biol 10(11): R128.
  7. Rasmussen, M., Sundstrom, M., Goransson Kultima, H., Botling, J., Micke, P., Birgisson, H., Glimelius, B. and Isaksson, A. (2011). Allele-specific copy number analysis of tumor samples with aneuploidy and tumor heterogeneity. Genome Biol 12(10): R108.
  8. Staaf, J., Vallon-Christersson, J., Lindgren, D., Juliusson, G., Rosenquist, R., Hoglund, M., Borg, A. and Ringner, M. (2008). Normalization of Illumina Infinium whole-genome SNP data improves copy number estimates and allelic intensity ratios. BMC Bioinformatics 9: 409.
  9. Van Loo, P., Nordgard, S. H., Lingjaerde, O. C., Russnes, H. G., Rye, I. H., Sun, W., Weigman, V. J., Marynen, P., Zetterberg, A., Naume, B., Perou, C. M., Borresen-Dale, A. L. and Kristensen, V. N. (2010). Allele-specific copy number analysis of tumors. Proc Natl Acad Sci U S A 107(39): 16910-16915.

简介

在乳腺癌和卵巢癌中发现了同源重组缺陷,主要由BRCA1或BRCA2灭活(所谓的BRCAness)引起。检测肿瘤中的BRCA1/2的实际失活对于患者的治疗和随访是重要的,因为它可以帮助预测对DNA损伤剂的应答并给出指示同源重组缺陷,主要是由于 > BRCA1 或 BRCA2 失活(所谓的BRCAness),可在乳腺癌和卵巢癌中发现。检测肿瘤中的BRCA1/2的实际失活对于基因测试是重要的。该协议描述如何基于由SNP阵列测量的肿瘤基因组谱来检测同源重组的损伤。建议的BRCAness的签名与在过滤和平滑小规模改变后计算的肿瘤基因组中大规模染色体断裂的数目有关。该程序强烈依赖于对绝对拷贝数和等位基因内容(等位基因特异性拷贝数)谱进行预处理的高质量SNP阵列。这种同源重组缺陷的基因组特征被证明在预测三阴性乳腺癌中的BRCA1/2灭活中是高度可靠的(97%的准确性;更多细节参见Popova等人 ,2012)和预测卵巢癌的存活(未发表数据)。作者感谢Dominique Stoppa-Lyonnet,Anne Vincent-Salomon,Thierry Dubois和Xavier Sastre-Garau的贡献。 (专利存放:参考编号EP12305648.3,2012年6月7日)

数据和软件

I.  数据:

  1. 肿瘤的全基因组SNP阵列谱。 Affymetrix和Illumina是提供用于初级标准化的高质量SNP阵列芯片和软件的主要平台。具体协议可在制造商的网站 www.affymetrix.com www.illumina.com。 SNP阵列谱需要通过某些软件进一步处理为绝对拷贝数和等位基因内容谱,用于挖掘SNP阵列谱。正确处理的SNP阵列谱的良好实例在Sanger研究所的癌细胞系收集中。
    http://www.sanger.ac.uk/cgi -bin/genetics/CGP/cghviewer/CghHome.cgi
  2. 从下一代测序(NGS)获得的SNP阵列样谱如果被处理成绝对拷贝数和等位基因内容谱,也可用于本方案;但是,我们在这里不详细考虑。


II。 软件:

  1. 用于SNP阵列初级标准化的软件:基因分型控制台,ChAS(两者都是 www.affymetrix.com ),Genome Studio ( www.illumina.com ),Aroma包(www.aroma-project.org ),用于Illumina阵列的分位数归一化的tQN(Staaf等人,2008)。br/>
  2. 用于挖掘SNP阵列谱以获得绝对拷贝数和等位基因内容(等位基因特异性拷贝数)的软件,例如GAP(Popova等人,2009),PICNIC(Greenman等人, ,2010),ASCAT(Van Loo等人,2010),GPHMM(Li等人,2011),TAPS(Rasmussen等人, al。,2011),Absolut(Carter ,2012)等。
  3. GAP方法和进一步的数据处理在R环境中实现( www.r-project.org )。但是,任何其他语言都可以用于执行此分析,包括MatLab,Java,C ++等。

设备

  1. 计算机(2 GHz,2 G RAM,Intel Core 2,40 G HD)

程序

I.   SNP阵列数据的预处理

  1. 根据阵列平台,通过适当的软件规范化.CEL文件。
  2. 将标准化数据导出到文本文件中:
    导出标准化数据:染色体,位置,Log_R_Ratio(Illumina)或Log_2_Ratio(Affymetrix),B等位基因频率(BAF,Illumina)或等位基因差异
  3. 通过(例如)GAP方法(Popova等人,2009)处理SNP阵列以获得(Birkbak等人,2011)正常污染的估计和(Carter ,2012)绝对拷贝数和等位基因内容概况(表1)。
    表1.分段的肿瘤基因组谱(来自Affymetrix OncoScan 300K的片段)
    位置开始
    位置结束
    染色体*
    长度SNP
    复制号码
    主要Allele
    59369
    115065314
    1
    13869
    2
    1
    115067829
    121049277
    1
    639
    3
    2
    143701096
    144299541
    1.5
    47
    3
    2
    144337336
    144989346
    1.5
    19
    0
    0
    145008423
    148551158
    1.5
    252
    3
    2
    148565769
    152700090
    1.5
    554
    6
    5
    * 1代表p臂,1.5代表染色体1的q臂; 中心区域用红色表示。


II。 质量控制

  1. 测量的SNP阵列轮廓的质量控制:用于初级数据归一化的软件通常提供每个芯片的质量指数;指示具有边缘质量的芯片必须从进一步分析中排除;可在相应的用户指南中找到质量截止的指示
  2. 通过正常基质细胞对肿瘤样品的污染:具有超过65%的预测正常细胞混合物的样品必须从进一步分析中排除(注意:测量的肿瘤样品通常代表不同比例的肿瘤和正常细胞的混合物,在测量的SNP阵列轮廓中导致不同的对比度; 60-70%的正常污染处于当前识别技术的极限)。
  3. 拷贝数和等位基因内容识别的质量控制:肿瘤基因组中拷贝数改变(CNA)的模式必须是"可解释的",意味着拷贝数变异和等位基因不平衡概况必须 始终如一。不幸的是,没有发展这种一致性的可靠测量;我们使用基于GAP图的手动识别控制(图1)。肿瘤基因组的GAP图是分段的SNP阵列谱的二维表示,其中每个圆代表一个片段(Popova等人,2009)。 GAP图的清晰和规则结构表明一致性(图1A),而混乱结构表明不一致(图1B)。具有不一致概况的样品必须从进一步分析中排除。


    图1.由Affymetrix OncoScan 300K测量的两种肿瘤样品的GAP图,表示(A)高质量图谱和(B)低质量图谱。肿瘤基因组的GAP图是分段SNP阵列的二维表示其中每个圆代表一个段(Popova等人,2009)。肿瘤样品来自GEO数据库(GSE28330,Birkbak等人,2012)。

  4. 充分检测染色体断裂的质量控制:高度污染的肿瘤样品连同SNP阵列谱中的非特异性变化通常导致通过分割算法检测的假阳性染色体断裂;在大量假阳性断裂的情况下,样品需要被丢弃。此类质量控制的适当正式程序尚未制定。我们对拷贝数变异和识别的拷贝数概况中检测到的断裂的一致性进行了粗略的视觉估计。
    注意:质量差的样品包括约10-15%的杂交样品,包括低肿瘤含量,低杂交,低识别质量等。


III。计算来自分段概况的大规模染色体断裂的数目(表1,图2):

注意:这里我们描述如何估计与同源重组缺陷相关的染色体断裂的数量;变化的过滤仅仅为了估计断点数目的目的而执行,并且与具体的变化无关,无论它们是什么重要的。

  1. 滤出微变异:微变异的大小S_micro是可检测的体细胞变化大小的下限,其取决于阵列中的SNP密度;例如,我们使用50个SNP用于Affymetrix SNP 6.0; 30个SNPs用于Illumina 600K;
    注意:此过滤的主要原因是微观变化通常与种系的拷贝数变异有关。
    1. 从分段的基因组谱中排除小于S_micro SNP的所有片段,并且如果它们具有相同的拷贝数和主要等位基因,则链接相邻片段。
  2. 过滤和平滑小规模变化:小规模变化的大小S_small < 3 Mb,定义在(Popova等人,2012)中。
    注意:该定义的主要原因是从3个Mb染色体断裂开始遵循泊松分布,即彼此独立;而小规模区段倾向于聚集在不连续的染色体区域。
    1. 根据大小订购小尺寸段。
    2. 从分段的基因组谱中排除最小的片段,并链接相邻片段(如果它们具有相同的复制数字和主要等位基因)。
    3. 重复过滤和平滑,直到最后一个小段。
      注意:过滤和平滑小规模变体的方式对生成的配置文件有轻微影响。
  3. 计算大小S Mb:LST_SMb的大规模状态转换(LST)的计数被定义为两个相邻(<3 Mb之间)段中的染色体断裂(拷贝数或等位基因含量的变化)> = S Mb每;在过滤和平滑小规模变异后,直接从分割的基因组谱中计算LST的数目(图2)。
    1. 标记染色体断裂如下:如果来自相同染色体臂的两个片段在拷贝数或主等位基因方面不同,并且大小为> = SMb, 3 Mb,中断标注为LST_SMb;
    2. 计算肿瘤基因组中LST_SMb(S = 3,4,...,11 Mb)的数目。
      注意:不考虑centromeric休息。


      图2.具有检测的拷贝数和LST的一条染色体的基因组谱的实例。 LRR:log R ratio profile; BAF:B等位基因频率分布; GT:由GAP识别的节段基因型; LST_10 Mb:黑色箭头指向检测到LST_10 Mb。黑色底线显示在过滤和平滑在GT轮廓中看到的小尺度变化之后获得的大尺度段。显示了来自GEO数据库的肿瘤样品的染色体3(GSE28330,Birkbak等人,2012)。


IV。 肿瘤倍性的估计:

  1. 估计肿瘤基因组的DNA指数,作为基因组中的平均拷贝数除以2.
  2. 根据以下规则,估计肿瘤基因组中的染色体计数作为每条染色体臂的中心区域拷贝数的总和(表1):
    1. 如果心室区域中的节段的大小为≥1.5Mb(或Affymetrix SNP6.0的500个SNP),则将相应染色体臂的拷贝数设置为该节段的拷贝数;
    2. 如果中心区域中的节段的大小为 1.5 Mb,染色体臂数由其平均拷贝数替代。
      注意:过滤微观变化后估计染色体数。
  3. 根据以下规则估计肿瘤倍性:如果DNA指数<1,则肿瘤倍性估计为2(近二倍体基因组) 1.3和染色体计数60;如果DNA指数≥1.3且染色体计数≥60,则估计肿瘤倍性为4(近四倍体基因组)。
    注意:这种归因是基于大量肿瘤基因组的分析获得的,用于乳腺癌和卵巢癌基因组(Popova等人,2012年, )。具有倍性的不明确归因的基因组在所考虑的所有病例中所占的比例小于5%。其他癌症可能有不同的基因组进化,倍数归因的阈值可能需要调整。


V. 肿瘤基因组中同源重组缺陷的标志

基于对具有已知状态的BRCA1/2基因的大系列乳腺癌的分析,发现LST_6,7,8,9,10Mb的数目代表了具有天然限定的倍性的有效鉴别特征特异性截止值,其允许以高精度和高精度预测BRCA1/2失活(表2)。测试卵巢癌的特征显示LST_6,7 Mb是与乳腺癌队列截止值相似的最有效的预测特征。

  1. 如果肿瘤基因组中LST的数目高于相应的倍性特异性截止值,则肿瘤基因组被注释为同源重组缺陷(表2)。
    注意:由于在基因组中检测到假阳性断裂,具有边界LST数目的肿瘤可能是假阳性。 LST_6,7,8,9,10 Mb预测之间的不一致性很少。
    表2.用于预测乳腺癌BRCAness的LST值的临界值
    LST_S Mb,S
    倍性2:(p = 68,N = 182)
    倍性4:(P = 53,N = 123)
    截止*
    FPR
    TPR
    切断
    FPR
    TPR
    6
    19(17)
    0.04
    0.99
    32(32)
    0.10
    1
    7
    17(15) 0.05
    0.99
    29(27)
    0.07
    0.98
    8
    14(14)
    0.06
    1
    26(26)
    0.08
    1
    9
    14(11)
    0.04
    0.99
    25(19)
    0.07
    0.98
    10
    11(11)
    0.07
    1
    22(18)
    0.06
    0.98
    *截断值对应于最大值(TPR-FPR);括号中的截止值对应于100灵敏度。
    P:BRCA1/2突变肿瘤的数目; N:BRCA1/2野生型或未测试肿瘤的数目;
    TPR:真阳性率; FPR:假阳性率。

参考文献

  1. Birkbak,NJ,Wang,ZC,Kim,JY,Eklund,AC,Li,Q.,Tian,R.,Bowman-Colin,C.,Li,Y.,Greene-Colozzi,A.,Iglehart,JD,Tung ,N.,Ryan,PD,Garber,JE,Silver,DP,Szallasi,Z.and Richardson,AL(2012)。 端粒等位基因不平衡表明DNA修复缺陷和对DNA损伤剂的敏感性。 Cancer Discov 2(4):366-375。
  2. Carter,SL,Cibulskis,K.,Helman,E.,McKenna,A.,Shen,H.,Zack,T.,Laird,PW,Onofrio,RC,Winckler,W.,Weir,BA,Beroukhim, ,Pellman,D.,Levine,DA,Lander,ES,Meyerson,M。和Getz,G。(2012)。 绝对定量人类癌症中体细胞DNA的改变。 Nat Biotechnol < em> 30(5):413-421。
  3. Greenman,CD,Bignell,G.,Butler,A.,Edkins,S.,Hinton,J.,Beare,D.,Swamy,S.,Santarius,T.,Chen,L.,Widaa, ,PA和Stratton,MR(2010)。 PICNIC:预测绝对等位基因拷贝数变异的算法与微阵列癌数据。 生物统计学 11(1):164-175。
  4. Li,A.,Liu,Z.,Lezon-Geyda,K.,Sarkar,S.,Lannin,D.,Schulz,V.,Krop,I.,Winer,E.,Harris, 。(2011)。 GPHMM:用于识别拷贝数改变和丢失的整合的隐马尔可夫模型使用全基因组SNP阵列在复杂肿瘤样品中的杂合性。核酸研究39(12):4928-4941。
  5. Popova,T.,Manie,E.,Rieunier,G.,Caux-Moncoutier,V.,Tirapo,C.,Dubois,T.,Delattre,O.,Sigal-Zafrani,B.,Bollet,M.,Longy ,M.,Houdayer,C.,Sastre-Garau,X.,Vincent-Salomon,A.,Stoppa-Lyonnet,D.and Stern,MH(2012)。 倍性和大规模基因组不稳定性始终使用BRCA1/2鉴定基底样乳腺癌 72(21):5454-5462。
  6. Popova,T.,Manie,E.,Stoppa-Lyonnet,D.,Rigaill,G.,Barillot,E.and Stern,M.H。(2009)。 基因组改变打印(GAP):用于可视化和挖掘由SNP阵列获得的复杂癌症基因组谱的工具。 Genome Biol 10(11):R128
  7. Rasmussen,M.,Sundstrom,M.,Goransson Kultima,H.,Botling,J.,Micke,P.,Birgisson,H.,Glimelius,B.and Isaksson,A。 具有非整倍体和肿瘤异质性的肿瘤样品的等位基因特异性拷贝数分析。 Genome Biol 12(10):R108
  8. Staaf,J.,Vallon-Christersson,J.,Lindgren,D.,Juliusson,G.,Rosenquist,R.,Hoglund,M.,Borg,A.and Ringner, _blank"href ="http://www.ncbi.nlm.nih.gov/pubmed/18831757"> Illumina的归一化全基因组SNP数据提高了拷贝数估计和等位基因强度比率。 BMC Bioinformatics 9:409.
  9. Van Loo,P.,Nordgard,SH,Lingjaerde,OC,Russnes,HG,Rye,IH,Sun,W.,Weigman,VJ,Marynen,P.,Zetterberg,A.,Naume,B.,Perou, Borresen-Dale,AL和Kristensen,VN(2010)。 肿瘤的等位基因特异性拷贝数分析。美国国家科学院院刊 107(39):16910-16915。
  • English
  • 中文翻译
免责声明 × 为了向广大用户提供经翻译的内容,www.bio-protocol.org 采用人工翻译与计算机翻译结合的技术翻译了本文章。基于计算机的翻译质量再高,也不及 100% 的人工翻译的质量。为此,我们始终建议用户参考原始英文版本。 Bio-protocol., LLC对翻译版本的准确性不承担任何责任。
Copyright: © 2013 The Authors; exclusive licensee Bio-protocol LLC.
引用:Popova, T., Manié, E. and Stern, M. (2013). Genomic Signature of Homologous Recombination Deficiency in Breast and Ovarian Cancers. Bio-protocol 3(13): e814. DOI: 10.21769/BioProtoc.814.
提问与回复

(提问前,请先登录)bio-protocol作为媒介平台,会将您的问题转发给作者,并将作者的回复发送至您的邮箱(在bio-protocol注册时所用的邮箱)。为了作者与用户间沟通流畅(作者能准确理解您所遇到的问题并给与正确的建议),我们鼓励用户用图片或者视频的形式来说明遇到的问题。由于本平台用Youtube储存、播放视频,作者需要google 账户来上传视频。

当遇到任务问题时,强烈推荐您提交相关数据(如截屏或视频)。由于Bio-protocol使用Youtube存储、播放视频,如需上传视频,您可能需要一个谷歌账号。