cluster

为什么关注聚类

k-means

原理

相关内容

• Start with suitable choices of K cluster mean-vectors. Then iterate the following two steps：
• For each object, find the closest mean vector and assign the the object to the corresponding cluster.
• For each cluster, update its mean vector according to the current assignments.
• Convergence is easy stop when cluster assignment no longer changes
• The choice of initial values is more difficult

Hierarchical Clustering

Produce a nested sequence of clusters, a tree, also called Dendrogram

结构变异 Genomic alterations

癌症中尤为明显

适合用 circos 来展示

SV

Mechanisms of SVs

• NAHR - non-allelic homologous recombination
• NHEJ - joining of DNA double-strand breaks without extensive sequence homology by ligation of DNA ends
• Breakage-fusion-bridge (BFB)
• Telomere loss or double-strand breakage creates an unprotected DNA end
• Ligation of broken chromatid ends connects the two sister chromatids
• Chromothripsis： A large number of genomic rearrangements have been acquired in a single catastrophic event

用高通量数据 short reads 来探测 SV

一些具体方法（可参考原始文献）

通过以下四个信息来进行预测

• read pairs
• read depth
• split reads
• and also assembly of the reads into contig

Copy number alterations （CV）

Use network flow to find out the traversal in the genome with CNAs

• Aneuploidy 非整倍体，染色体中出现非整倍体的变化，癌症中很常见

• Quantify allele-specific SVs and CNAs 软件：Weaver

---

本文作者：思考问题的熊

版权声明：本博客所有文章除特别声明外，均采用 知识共享署名-非商业性使用-禁止演绎 4.0 国际许可协议 (CC BY-NC-ND 4.0) 进行许可。

如果你对这篇文章感兴趣，欢迎通过邮箱或者微信订阅我的 「熊言熊语」会员通讯，我将第一时间与你分享肿瘤生物医药领域最新行业研究进展和我的所思所学所想，点此链接即可进行免费订阅。

· 分享链接 https://kaopubear.top/blog/2017-08-12-longxing-bioinfo-cluster/