${cell}_mapQ30_sort_rmdup.bed.gz ->

[Sun-Wu-Kong-666]

when I get a .bed file shown below, then how can I get a _SegCopy file(i.e. bin.info) shown below?

/lustre1/tangfuchou_pkuhpc/Software/bedtools2/bin/bamToBed -i ${cell}_mapQ30_sort_rmdup.bam |gzip > ${cell}_mapQ30_sort_rmdup.bed.gz

bin.info <- read.table("data/SegCopy/DNA_A1_lym_1_FKDL210056832-1a_SegCopy",
header = T,row.names = bin_name)[,1:3]

[liuzhenyu-yyy]

The _SegCopy file is a matrix that shows the estimated copy number of each single cell in each genomic bin.

There are various algorithms available for inferring single-cell copy number variations using single-cell whole genome sequencing data. For our study, we used Ginkgo (https://github.com/robertaboukhalil/ginkgo), an open-source web platform that is user-friendly and was published in Nature Methods (https://www.nature.com/articles/nmeth.3578).

The details of this pipeline are summarized in the "Materials and methods" section of our publication.

Estimation of single-cell CNAs from single-cell genome sequencing data
Single-cell CNAs were estimated as described in a previous study [35]. The genome was tiled into 2808 variable-sized bins with an approximate length of 1.0 Mb. Predefined ‘bad bins’ were excluded from further processing. Read counts in each bin were normalized by sequence depth of the cell and then corrected for GC content using LOWESS regression. Cells from the same 96-well plates were pooled together as a control to eliminate amplification bias. Circular binary segmentation (CBS) was used to segment the read count profile of each cell, with the parameters ‘alpha = 1E−10’ and ‘undo.prune = 0.05’. Copy number of each segment was then determined using numerical optimization. For each cell, the coefficient of variation (CV) of read count in each bin was calculated. Cells with CV ≥ 1 or total usable unique read number ≤ 3E−5 were excluded. Remaining cells were further filtered for CNA profiles by an additional manual check. In total, 1616 of 2208 cells (73.2%) passed quality control and were retained in downstream analysis. Only cells with CNAs longer than 10 Mb and confirmed to be of good quality by manual check were considered to have CNAs. For female cells, the Y chromosome was not considered.

Please let me know if you have any further questions or if there's anything else I can help you with.

[Sun-Wu-Kong-666]

非常感谢您的回复。
请问您是在Ginkgo网页(http://qb.cshl.edu/ginkgo/?q=/Wd5WdknaXLVY63CMYgg4) 将bed文件上传后进行分析的吗？还是在本地安装了Ginkgo后进行分析的，我尝试在Ginkgo网页上传了bed文件，但中途报错，无法完成。
另外您有其他专门分析单细胞全基因组测序数据的工具推荐的吗？我看了一些流程，使用CNVnator，但它并非专门分析单细胞的，担心不够准确。

[liuzhenyu-yyy]

Estimation of single-cell copy number aberrations (CNAs) from single-cell genome sequencing data was performed locally on our high-performance computing platform.

There are many different software options available for this goal, including Control-FREEC, HMMcopy, and CHISEL. These methods have been summarized and benchmarked in multiple research studies. One such study, entitled ‘Methods for copy number aberration detection from single-cell DNA-sequencing data’ on Genome Biology, reviewed eight methods that have been developed for detecting CNAs in scDNAseq data and categorized them according to the steps of a seven-step pipeline that they employ.

Similar articles should be easily available through Google or PubMed.

[Sun-Wu-Kong-666]

Thanks very much!