tadesouaiaia
commented
3 days ago Yes you are correct. Would you mind sending us the whole command line printout or the standard errors from inside the prs-single directories?
Thank you,
Tade
On Tue, Oct 22, 2024 at 12:46 PM mpkol59 @.***> wrote:
Hello there, Thank you for your great work on this software. I have encountered this error while trying to run BridgePRS using my East Asian and European ancestry data. My command line is as found below as well as the error message:
./bridgePRS tools check-pops -o mytestfulleas --pop EAS EUR --ld_path /path/AFR/CHRMS/1000G/ --sumstats_prefix /path/EAS/CHRMS/eas_chr /path/AFR/CHRMS/eur_chr --sumstats_size 1600 390278 --genotype_prefix /path/EAS/CHRMS/ceas_chr --phenotype_file /path/EAS/CHRMS/myeasphenos2.data
../../bridgePRS pipeline go -o out2 --config_files EAS.target.config EUR.base.config --phenotype y --debug_level 0
################################################################# Traceback (most recent call last): File "/path/BridgePRS/mytestfulleas/save/../../bridgePRS", line 372, in bridgePRS = BridgePRS(mps, sys.argv) ^^^^^^^^^^^^^^^^^^^^^^^^ File "/path/BridgePRS/src/Python/BridgePRS.py", line 27, in init elif self.args.module == 'pipeline': self.easyrun() ^^^^^^^^^^^^^^ File "/path/BridgePRS/src/Python/BridgePRS.py", line 46, in easyrun self.analyze(self.args.cmd, self.args.result_files, PATH = self.io.paths['home'])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/path/BridgePRS/src/Python/BridgePRS.py", line 69, in analyze self.run = BridgeRun(self).analyze(cmd, prs_results, PATH) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/path/BridgePRS/src/Python/Util/BridgeRun.py", line 65, in analyze self.results = [r1[0], r2[0]] + BridgeResult().read_combo(combine) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/path/BridgePRS/src/Python/Util/Bridge_Run/BridgeResult.py", line 40, in read_combo return [BridgeResult('prs-combine', combo.pop, True).add_triple([varexp[0], preds[0], snp_weights[0]]), BridgeResult('prs-weighted', combo.pop, True).add_triple([varexp[1], preds[1], snp_weights[1]])] ^^^^^^ UnboundLocalError: cannot access local variable 'varexp' where it is not associated with a value
##################################################################################
This is confusing since I ran the exact same command for my African and Eur populations and did not get this error. What I noticed is that the PRS-Single stage runs into some error at the beta estimation stage and just continues on to the final stage where the needed files for combining the models are then found missing, hence the error.
I would really appreciate your help with resolving this. Thanks as I look forward to hearing back from you.
Best wishes.
— Reply to this email directly, view it on GitHub https://github.com/clivehoggart/BridgePRS/issues/6, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABRCX7TT4RFRUTOV3GYX4WTZ4Z6OFAVCNFSM6AAAAABQM53UZGVHI2DSMVQWIX3LMV43ASLTON2WKOZSGYYDKOJXGM4DMNI . You are receiving this because you are subscribed to this thread.Message ID: @.***>
mpkol59
Hello there, Thank you for your great work on this software. I have encountered this error while trying to run BridgePRS using my East Asian and European ancestry data. My command line is as found below as well as the error message:
./bridgePRS tools check-pops -o mytestfulleas --pop EAS EUR --ld_path /path/AFR/CHRMS/1000G/ --sumstats_prefix /path/EAS/CHRMS/eas_chr /path/AFR/CHRMS/eur_chr --sumstats_size 1600 390278 --genotype_prefix /path/EAS/CHRMS/ceas_chr --phenotype_file /path/EAS/CHRMS/myeasphenos2.data
../../bridgePRS pipeline go -o out2 --config_files EAS.target.config EUR.base.config --phenotype y --debug_level 0
################################################################# Traceback (most recent call last): File "/path/BridgePRS/mytestfulleas/save/../../bridgePRS", line 372, in
bridgePRS = BridgePRS(mps, sys.argv)
^^^^^^^^^^^^^^^^^^^^^^^^
File "/path/BridgePRS/src/Python/BridgePRS.py", line 27, in init
elif self.args.module == 'pipeline': self.easyrun()
^^^^^^^^^^^^^^
File "/path/BridgePRS/src/Python/BridgePRS.py", line 46, in easyrun
self.analyze(self.args.cmd, self.args.result_files, PATH = self.io.paths['home'])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/path/BridgePRS/src/Python/BridgePRS.py", line 69, in analyze
self.run = BridgeRun(self).analyze(cmd, prs_results, PATH)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/path/BridgePRS/src/Python/Util/BridgeRun.py", line 65, in analyze
self.results = [r1[0], r2[0]] + BridgeResult().read_combo(combine)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/path/BridgePRS/src/Python/Util/Bridge_Run/BridgeResult.py", line 40, in read_combo
return [BridgeResult('prs-combine', combo.pop, True).add_triple([varexp[0], preds[0], snp_weights[0]]), BridgeResult('prs-weighted', combo.pop, True).add_triple([varexp[1], preds[1], snp_weights[1]])]
^^^^^^ UnboundLocalError: cannot access local variable 'varexp' where it is not associated with a value ##################################################################################
This is confusing since I ran the exact same command for my African and Eur populations and did not get this error. What I noticed is that the PRS-Single stage runs into some error at the beta estimation stage and just continues on to the final stage where the needed files for combining the models are then found missing, hence the error.
I would really appreciate your help with resolving this. Thanks as I look forward to hearing back from you.
Best wishes.