search

BernsteinLab / Myeloid-Glioma

Scripts used to perform the analyses to study the Biology of Myeloid Cells in Glioma Microenvironments
GNU General Public License v3.0
4 stars 0 forks source link

Usage matrix value is NA #2

Open weiju455 opened 2 weeks ago

weiju455 commented 2 weeks ago

Hello, I like your article very much, I have a problem, why my usage matrix value is NA, and then the gene matrix is normal 屏幕截图 2024-10-23 103639

tymiller commented 2 weeks ago

Is this after using the shiny app or trying to run cNMF locally?

Tyler E. Miller, MD, PhD Assistant Professor of Pathology Case Western Reserve University School of Medicine Molecular Pathologist University Hospitals Cleveland Medical Center Member, Case Comprehensive Cancer Center tymillerlab.orghttps://tymillerlab.org/

From: weiju455 @.> Sent: Wednesday, October 23, 2024 2:27:23 AM To: BernsteinLab/Myeloid-Glioma @.> Cc: Subscribed @.***> Subject: [BernsteinLab/Myeloid-Glioma] Usage matrix value is NA (Issue #2)

Hello, I like your article very much, I have a problem, why my usage matrix value is NA, and then the gene matrix is normal 2024-10-23.103639.png (view on web)https://github.com/user-attachments/assets/5aaa03ee-58e5-4409-8a1e-1a2c591e7926

— Reply to this email directly, view it on GitHubhttps://github.com/BernsteinLab/Myeloid-Glioma/issues/2, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AAIQHAEZT2U2WQOQTGJDZGLZ446UXAVCNFSM6AAAAABQOBF2JWVHI2DSMVQWIX3LMV43ASLTON2WKOZSGYYDONBSHEYTSOI. You are receiving this because you are subscribed to this thread.Message ID: @.***>

weiju455 commented 2 weeks ago

Thank you so much for getting back to me. I am run cNMF locally.But I don't have this error when I run each dataset individually, I put it together and I take the first 2000 high-variable genes and I run it separately and I have this error.Also, there are too many zeros in my matrix, and it seems to be too sparse.

tymiller commented 2 weeks ago

I'll let Chadi comment on the local version. I'm not sure what would cause this issue.

On Wed, Oct 23, 2024 at 8:27 AM weiju455 @.***> wrote:

Thank you so much for getting back to me. I am run cNMF locally.But I don't have this error when I run each dataset individually, I put it together and I take the first 2000 high-variable genes and I run it separately and I have this error.Also, there are too many zeros in my matrix, and it seems to be too sparse.

— Reply to this email directly, view it on GitHub https://github.com/BernsteinLab/Myeloid-Glioma/issues/2#issuecomment-2431979036, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAIQHAG4JSUQMWZFX5RYUBTZ46I3BAVCNFSM6AAAAABQOBF2JWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMZRHE3TSMBTGY . You are receiving this because you commented.Message ID: @.***>

--

Tyler E. Miller, MD, PhD Paul and Betsy Shiverick Professor of Immuno-Oncology Assistant Professor of Pathology Case Western Reserve University School of Medicine Molecular Pathologist University Hospitals Cleveland Medical Center Member, Case Comprehensive Cancer Center tymillerlab.org

chadielfarran commented 2 weeks ago

Hi weiju455,

Could you give me more details to help you address the issue?

1- What does your input look like (Could you screenshot a few lines from the input)? 2- Is cNMF issuing any error messages or warnings?

weiju455 commented 2 weeks ago

Hi chadielfarran,Thank you for your reply. 1,This is my input ,This is the matrix that ran the problem， image GBM07_filtered_matrix.txt This is the matrix that runs through without any problems,but But this number of genes is much larger than the 2,000 defined in the cnmf parameters，This file is too large to upload image 2、No errors or warnings were reported during the cnmf process 3、At one point, I ran the cnmf on each data (they all had a large number of genes) and then crossed the genes in spectra and found that the number was small. Then I first combined 3000 high-variant genes (2000 to start with, problem. Then, after filtering the dataset, each data gene becomes 3000, and then running the cnmf, it is found that the value in usage is NA

chadielfarran commented 2 weeks ago

Thank you for providing the details. It could be a matrix merging issue. May I know how you merged all the datasets?

Another possibility is having a column (gene) whose expression values are 0 across all the cells (which could possibly happen during the merging of the matrices).

Could you check if any gene has a sum of 0 in all the cells?