Introduction to single cell ATAC data analysis in R
HTML-код
- Опубликовано: 2 окт 2022
- This is a primer for single cell/nuclei ATAC-seq data analysis. What is single cell ATACseq? How do you perform basic scATAC-seq analysis in R? I describe what scATACseq is. Then I use Seurat and Signac to do data analysis using a recent Nature communications paper. I do preprocessing, clustering, differential accessibility analysis, RNA activity estimation, and I make various plots.
Notebook:
github.com/mousepixels/sanbom...
References:
stuartlab.org/signac/index.html
satijalab.org/seurat/articles...
www.nature.com/articles/s4146... 
Наука
![[WEBINAR] Analysis of Single-Cell Multiome ATAC + Gene Expression - Dr. Wayne Doyle](http://i.ytimg.com/vi/y1mFdkDVc-c/mqdefault.jpg)
Thank you so much for your all great videos. Truly appreciate it.
No problem!
thank you for this video. You always make everything simple. So great! 😁
You are so welcome! :)
Thank you so much for such an amazing tutorial!!
Glad it was helpful!
Great video! I truely got inspired! Could you make videos about Chipseq and proteomics data analysis if it is possible?
Hi, I was wondering what to use in the "genome" bit when doing the CreateChromatinAssay I have a custom reference genome that I am using so it is not available on any of the online databases. Thank you!
Thank you for such a wonderful video. I have a one question, how you fixed the error of import_atac (import_atac function not found)
Thank you always!
Thanks for watching and leaving a comment!
Thank you for the video. Is the order of RunTFIDF and FindTopFeatures in the video opposite? first normalyze than find the top features, right? Regarding latent.vars = 'peak_region_fragments', can you please elabrate the use of peak_region_fragments as a covarient?
Hi Sanbomics,
If I only have fragment files in the dataset I want to use, how should I proceed?
Thank you for the video primer. Question: your UMAP for the combined dataset shows the young separate from the old. How do you know this is biological and not a batch effect?
Ahh the age old question. It's hard to say, but likely some of both. One thing I could (should) have done is run cellranger agg on the samples. But I doubt it would have made the two subsets overlap entirely. You could try integration, but then you run into the possibility of erasing true biological variability. Still one of the most prominent issues in the field
Thank you so much for this video! I keep coming to it all the time during my analysis. I have a one question though, how is it that in all my samples I do not seem to have 'blacklist_region_fragments'? I mean, I have a column but the values are all '0'. And it is the case with all the samples I have... Everything else seems to work fine and I can follow the whole tutorial but I never get to see the 'blacklist_ratio' and cannot filter it out. Thank you 🙏
No problem! Hmm, were the data you used 10x libraries processed with 10x software? I thought it would be included by default in the metadata. Don't sweat it too much though. I have processed other data that don't have them just fine.
Yes, it was processed by CellRanger so not sure why I don't have them 🤔 not going to worry about it though 🙂 Thanks! 💐
Strange.. maybe an older version or something. Not sure. If you have access to fastq you can try rerunning using the most recent ref/software version
thank you so much for all those amazing vidios, i am soon starting my Bioinformatic internship its all about analysis of SC data so this is so helpfull for me.
i just want to take ur advice about something, i all my life was a wetlab guy, now switching to the computers isn't that easy, i find my self understanding the steps and workflow but am not comfortable yet with the coding and i spend alot of time trying understand the meaning of each code you use, so i wanted to ask if you have any recommendation for playlists or video's to help me speed up my coding skills in python and r ?
is there a method to follow or any advice? thanks alot again for all this amazing video's
Thats how I started years ago. My best advice is come up with a project. It doesn't even have to be related to bioinformatics. It could be something like writing a program that can play a simple card game that you like. It doesn't matter what it is, but I feel like having a goal is the best way to start. Then only watch videos and tutorials when you are trying to learn something very specific. Writing code to do something helps you learn faster than trying to remember someone else's code in my opinion.
🎉🎉🎉
Wonderful and informative video. But can you do the whole tutorial video like you did previously on rna seq database ?
I am currently doing project with atac seq data but still struggling.
I plan to eventually! I've been very busy irl though, sadly
@@sanbomics if you can make one, i really appreciate. Your videos are so helpful to me and those who are new to this field. Thank you so much 😊! Hope you can make more videos frequently in the future .
Hi Mark, I have ATAC seq data in these file format: bam files, .bw files, and fastq files. I don't have .h5 , csv and tsv file as you did in this tutorial. So I am trying to install some ATAC pipeline which is a lot more complicated, would you please have a suggestion in my case?
What technology was used to generate it?
@@sanbomics From 10x genomics I think.
@@sanbomics Looking forward to hear from you 😅.
Is there a way to be in contact with you rather than Twitter and RUclips ? I would be grateful ! Thank you!
Umm, I don't really have a mechanism set up other than those. Github also I suppose. But twitter is the only place you can DM me.
thank you so much for this video, please how we can contact with you ?
You can contact me on Twitter if you have one!
@@sanbomics please write your Twitter account
its @sanbomics too
@@sanbomics okay thank you dear
What about .bw files?
Hi! What about them?