Hello, could you provide more information on using LDSC for cell type-specific analyses? Recently, I also need to use LDSC-SEG, so I searched for some literature and found that the results in https://www.nature.com/articles/s41588-022-01069-0#code-availability were very significant. Therefore, I would like to analyze the data with the code provided in the article. However, I couldn't obtain p-values that reached the level mentioned in the paper. I'm not sure if there were any errors in my procedure：
- I used Top coronary artery cell type snATAC marker peaks in the Supplementary Data 1.
- I used 1000G_Phase3_baseline_v1.2_ldscores.tgz and 1000G_Phase3_weights_hm3_no_MHC.tgz from https://zenodo.org/record/8292725.
- I run the code privided in the /4_RASQUAL_caQTL/ldsc folder.
- I used the Cahoy.control annotation file in the Cell-type-specific-analyses analysis to calculate the control LD scores.
I only run the CAD with cell type peaks, I got:
However, there is a significant disparity between the results I obtained and those provided in the article.
Is there any specific issue that might have caused this difference? Could you please assist me?
Hello, could you provide more information on using LDSC for cell type-specific analyses? Recently, I also need to use LDSC-SEG, so I searched for some literature and found that the results in https://www.nature.com/articles/s41588-022-01069-0#code-availability were very significant. Therefore, I would like to analyze the data with the code provided in the article. However, I couldn't obtain p-values that reached the level mentioned in the paper. I'm not sure if there were any errors in my procedure：
I only run the CAD with cell type peaks, I got:
However, there is a significant disparity between the results I obtained and those provided in the article.
Is there any specific issue that might have caused this difference? Could you please assist me?