expression matrices and high MOI data

[xujin...@gmail.com]

Dear Atray,

This study is very helpful to understand biological dynamics with your data. I have some questions about the resource.

(1) As compiled gene expression matrices have been provided in GSE90063, I think they can be served as input of MIMOSCA or just differential analysis. But I don't know if I can calculate the technical covariates and cell states just depend on these data.

(2) In GSE90063, the high MOI K562 TFs data is provided in two conditions: lenient and strict. What is the difference between them? By the way, is the data in Table s4 the low MOI K562 TFs data with lenient condition? Can you provide me the coefficient matrices of high MOI K562 TFs data?

Thank you very much!

Best regards!

Jinyuan

[Atray Dixit]

Dear Jinyuan,

1) See the post I just sent David.

"For the cell state covariate, see pg. e7 in our STAR methods section. Sections titled ("Definition of cell states", and "Relation of perturbed cells to unperturbed states") and the associated github example (https://github.com/asncd/MIMOSCA/tree/master/Cell_States) 

 

For the sgRNA covariates, see pg. e6 ("Alignment of cell barcode / GBC libraries") and the associated github example (https://github.com/asncd/MIMOSCA/tree/master/GBC_CBC_pairing) "

2) The difference is described on page e5, see below:

For strict filtering, chimeric reads were removed by thresholding molecules that, for a given unique combination of cell barcode and UMI, received less that 20% of the reads.

See https://github.com/asncd/MIMOSCA for more on filtering chimeric reads. 

and yes you're right about table s4. I can send you the matrices early next year.

I hope that helps!

Atray

[xujin...@gmail.com]

Thank you for your reply! I'm sorry for my unclear statement about the second question. Actually, I want to know if I use the high MOI data to identify individual relations between perturbed TFs and downstream genes, which would be better to select?

I'll be very happy if you could send the matrices :)

Best,

Jinyuan

[Atray Dixit]

Hi Jinyuan,

The strict is probably better. Just clarifying, the reason you want to look at the high MOI data is to look for interactions between TFs on their target genes? I'm just about to look for the files.

Best,

Atray

[xujin...@gmail.com]

Hi Atray,

Yes, I want to know if perturbations of multiple genes could be better to describe the interactions between TFs and their target genes. I think it could be helpful to understand the regulatory interactions from the perturbations of single gene and multiple genes.

Thank you for your help!

[Atray Dixit]

Hi Jinyuan,

Here are the matrices. I hope they help!

https://sendit.broadinstitute.org/files/download/acdixit/5a5e81018b3dc/package.zip

Best,

Atray

[xujin...@gmail.com]

Hi Atray,

Thank you very much! They are really helpful! :)

Best,

Jinyuan