SHARE-seq mouse skin velocyto output

[chelseabright96]

Good day,
I am interested to run MultiVelo on the SHARE-seq mouse skin dataset and compare the results with another model. I would like to request the unspliced and spliced count matrices that you used for your experiments if you still have them available. If not, It would be a huge help if you could share with me the input files you used for velocyto and the command for running velocyto. I am new to the tool and would appreciate any help as I am not sure about what files to use from the genome omnibus website.

Thanks in advance for your help!

[danielee0707]

I can share the Velocyto loom file through email if you share the address with me.

[chelseabright96]

Thank you! I have sent you an email

[prauten]

Thank you for developing MultiVelo. It's an excellent tool that we actively use in our projects.

I am currently designing final class projects for students and would love to incorporate scRNA-seq velocity estimations for the SHARE-seq dataset. To this end, I have been querying the data you have kindly provided as part of your manuscript and tutorials. For the SHARE-seq hair follicle differentiation subset, your preprocessed files - including spliced and unspliced counts along with estimates for Ms and Mu from scvelo- provide an excellent starting point for the student projects.

However, I have a follow-up question regarding the provided data. The processed data subset contains 6,436 cells, while the original data from Ma et al. on GEO lists 7,197 cells when I subset to the same cell populations (['TAC-1', 'IRS', 'Medulla', 'TAC-2', 'Hair Shaft-cuticle.cortex']). Did you perform any additional quality control that might have led to the removal of some cells from the original publication? Additionally, does the loom file you shared previously contain all cells from the Ma et al. mouse skin dataset? If you conducted any additional filtering that I may have overlooked in your manuscript, I would appreciate any information on that. If the loom file contains all cells, would you be so kind as to share it with me?

Thank you very much for your assistance!
Pia

[danielee0707]

Hi @prauten, the 6436 cells contain all the cells the original authors used for pseudotime analysis with Palantir. We requested and obtained a text file called ptime.bed directly from the authors. I believe that if you email the first author of SHARE-seq paper Dr. Sai Ma, he will kindly share the same file with you. And good luck with the class project design.

[prauten]

Hi @danielee0707, thank you very much for your swift response and help! I'll follow up with the SHARE-seq paper authors then. Kind regards, Pia.