Unexpected error: Execution failed on sql 'SELECT gene, phi FROM extra': no such column: phi

[kys21207]

I got the unexpected error like the below.
INFO - Started metaxcan association
INFO - 10 % of model's snps found so far in the gwas study
INFO - 20 % of model's snps found so far in the gwas study
INFO - 30 % of model's snps found so far in the gwas study
INFO - 40 % of model's snps found so far in the gwas study
INFO - 50 % of model's snps found so far in the gwas study
INFO - 60 % of model's snps found so far in the gwas study
INFO - 70 % of model's snps found so far in the gwas study
INFO - 80 % of model's snps found so far in the gwas study
INFO - 80 % of model's snps used
INFO - Calibrating pvalue and zscore using phi in model, N and h2
INFO - Unexpected error: Execution failed on sql 'SELECT gene, phi FROM extra': no such column: phi

python /app/MetaXcan/software/SPrediXcan.py --model_db_path transitional_stage_B_cell.db --gwas_file dx_uc_v3.input.txt.gz --snp_column rsid --beta_column BETA --effect_allele_column ALLELE1 --non_effect_allele_column ALLELE0 --zscore_column Z --covariance transitional_stage_B_cell_covariances.txt.gz --keep_non_rsid --pvalue_column P --gwas_h2 0.1689 --gwas_N 14012 --output_file dx_uc_v3.preixcan_transitional_stage_B_cell_cells.csv

[hakyim]

Festus, can you check this? Backwards compatibility issues?

…

On Thu, Feb 13, 2025 at 3:49 PM Kijoung Song ***@***.***> wrote: I got the unexpected error like the below. INFO - Started metaxcan association INFO - 10 % of model's snps found so far in the gwas study INFO - 20 % of model's snps found so far in the gwas study INFO - 30 % of model's snps found so far in the gwas study INFO - 40 % of model's snps found so far in the gwas study INFO - 50 % of model's snps found so far in the gwas study INFO - 60 % of model's snps found so far in the gwas study INFO - 70 % of model's snps found so far in the gwas study INFO - 80 % of model's snps found so far in the gwas study INFO - 80 % of model's snps used INFO - Calibrating pvalue and zscore using phi in model, N and h2 INFO - Unexpected error: Execution failed on sql 'SELECT gene, phi FROM extra': no such column: phi python /app/MetaXcan/software/SPrediXcan.py --model_db_path transitional_stage_B_cell.db --gwas_file dx_uc_v3.input.txt.gz --snp_column rsid --beta_column BETA --effect_allele_column ALLELE1 --non_effect_allele_column ALLELE0 --zscore_column Z --covariance transitional_stage_B_cell_covariances.txt.gz --keep_non_rsid --pvalue_column P --gwas_h2 0.1689 --gwas_N 14012 --output_file dx_uc_v3.preixcan_transitional_stage_B_cell_cells.csv — Reply to this email directly, view it on GitHub <#206>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAW2ROMELBFVEORTSHQ3JFD2PUHNHAVCNFSM6AAAAABXDG674SVHI2DSMVQWIX3LMV43ASLTON2WKOZSHA2TEMJXGU3DSMQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> [image: kys21207]*kys21207* created an issue (hakyimlab/MetaXcan#206) <#206> I got the unexpected error like the below. INFO - Started metaxcan association INFO - 10 % of model's snps found so far in the gwas study INFO - 20 % of model's snps found so far in the gwas study INFO - 30 % of model's snps found so far in the gwas study INFO - 40 % of model's snps found so far in the gwas study INFO - 50 % of model's snps found so far in the gwas study INFO - 60 % of model's snps found so far in the gwas study INFO - 70 % of model's snps found so far in the gwas study INFO - 80 % of model's snps found so far in the gwas study INFO - 80 % of model's snps used INFO - Calibrating pvalue and zscore using phi in model, N and h2 INFO - Unexpected error: Execution failed on sql 'SELECT gene, phi FROM extra': no such column: phi python /app/MetaXcan/software/SPrediXcan.py --model_db_path transitional_stage_B_cell.db --gwas_file dx_uc_v3.input.txt.gz --snp_column rsid --beta_column BETA --effect_allele_column ALLELE1 --non_effect_allele_column ALLELE0 --zscore_column Z --covariance transitional_stage_B_cell_covariances.txt.gz --keep_non_rsid --pvalue_column P --gwas_h2 0.1689 --gwas_N 14012 --output_file dx_uc_v3.preixcan_transitional_stage_B_cell_cells.csv — Reply to this email directly, view it on GitHub <#206>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAW2ROMELBFVEORTSHQ3JFD2PUHNHAVCNFSM6AAAAABXDG674SVHI2DSMVQWIX3LMV43ASLTON2WKOZSHA2TEMJXGU3DSMQ> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[Fnyasimi]

@kys21207 If you want to correct for the polygenicity induced inflation you can download the latest models with phi value from here with the tutorial explained here.

To run the normal SPredixcan as before just remove the --gwas_h2 and --gwas_N flag as they are only required when correcting for the polygenicity induced inflation.

[kys21207]

Thank you for your comments. It is very helpful. However, I have two questions in terms of the polygenicity induced inflation .

1. Should we always correct for inflation due to polygenicity? Are there any guidelines on when we should correct?
2. I am very interested in scpredixcan with immune cell types of oneK1K. Is there an updated immune cell type with phi values?