Error in impGa: C++ exception (unknown reason)

[wjiang-datasci]

Hi Zhili,

I tried to run SBayesRC with my own summary statistics and the Eur_HM3_UKBB LD reference provided by you: Rscript -e "SBayesRC::impute(mafile='${out_prefix}_tidy.ma', LDdir='$ld_folder', \ output='${out_prefix}_imp.ma', log2file=TRUE)"

However, I keep having Error message in impGa. The log message is attached below. I tried to run it on both our cluster equipped with slurm manager (OS: Red Hat Enterprise Linux 8, 8.6) and my own macbook (MacOS: 13.5.1), and I tried to run different versions of SBayesRC (from 1.2 to 2.0, and the developing version). The error message is the same for all attempts.

Could you enlighten me how to solve this problem?

Best,

Wei

Impute the summary data by LD 252905 SNPs in common between GWAS summary and LD 146895 SNPs set from summary data 106010 SNPs flipped alleles 252905 SNPs are typed SNPs Start summary imputation... ==========1========= Imputing... Read error (m) Error in impGa(info$template, idxBlk, info$type, ma_exist$z, idxtt, m, : c++ exception (unknown reason) Calls: -> impGa Execution halted

[zhilizheng]

Hi @wjiang-datasci ,

Thanks for the report and the nice detalis. I just download everything from the link I shared, setup a new system (RHEL 8.8 and 7.6) and have tests, I can't replicate the problem. From the error log, it looks like your LD file is broken, could you re-download the LD file and unzip again? (please keep the R package ver >= 0.2.0, old versions would not work)

wget https://plot.cnsgenomics.com/SBayesRC/LD/HapMap3/ukbEUR_HM3.zip
unzip ukbEUR_HM3.zip

Don't hesitate to let me know if this doesn't solve the problem.

Regards, Zhili

[wjiang-datasci]

Hi Zhili,

Thank you for your prompt reply. I have tried to download the LD file and unzip again. Still, the same issue happened. Is there other possible reason contributing to this issue?

Best,

Wei

[zhilizheng]

Hi @wjiang-datasci ,

Thanks for the following up. Did you check the package version? It looks like to me, that you are running an old version R package. Could you try to run this again:

install.packages("https://github.com/zhilizheng/SBayesRC/releases/download/v0.2.0/SBayesRC_0.2.0.tar.gz",repos=NULL, type="source")

Then, I guess it would go through, and you will meet another error. Let me know if this go through, and I will follow up with that error later.

Regards, Zhili

[wjiang-datasci]

Hi Zhili,

Yes, I installed the latest version:

packageVersion('SBayesRC') [1] ‘0.2.0’

And I have installed the package again, but still facing the same issue.

Best,

Wei

[zhilizheng]

Hi @wjiang-datasci

Thanks for the information. Could you paste the full log here ? I will look into this right now. Besides, could you share me the results of

lscpu
uname -r

Many thanks.

Regards, Zhili

[wjiang-datasci]

Sure, the log I pasted before is the full log I obtained. I also pasted it below with the results of lscpu and uname.

---

(SBayesRC)[wj@r206u22n12.mccleary SBayesRC]$ cat ~/result/SBayesRC/test_imp.ma.log Impute the summary data by LD 252905 SNPs in common between GWAS summary and LD 146895 SNPs set from summary data 106010 SNPs flipped alleles 252905 SNPs are typed SNPs Start summary imputation... ==========1========= Imputing... Read error (m) Error in impGa(info$template, idxBlk, info$type, ma_exist$z, idxtt, m, : c++ exception (unknown reason) Calls: -> impGa Execution halted

---

lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 46 bits physical, 48 bits virtual Byte Order: Little Endian CPU(s): 28 On-line CPU(s) list: 0-27 Vendor ID: GenuineIntel Model name: Intel(R) Xeon(R) CPU E5-2660 v4 @ 2.00GHz CPU family: 6 Model: 79 Thread(s) per core: 1 Core(s) per socket: 14 Socket(s): 2 Stepping: 1 BogoMIPS: 3999.78 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pa t pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes6 4 monitor ds_cpl smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsa ve avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat _l3 cdp_l3 invpcid_single intel_ppin ssbd ibrs ibpb stibp fsgsba se tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm rdt_a rdseed adx smap intel_pt xsaveopt cqm_llc cqm_occup_llc cqm_mbm _total cqm_mbm_local dtherm ida arat pln pts md_clear flush_l1d Caches (sum of all): L1d: 896 KiB (28 instances) L1i: 896 KiB (28 instances) L2: 7 MiB (28 instances) L3: 70 MiB (2 instances) NUMA: NUMA node(s): 2 NUMA node0 CPU(s): 0-13 NUMA node1 CPU(s): 14-27 Vulnerabilities: Itlb multihit: KVM: Mitigation: VMX unsupported L1tf: Mitigation; PTE Inversion Mds: Vulnerable; SMT disabled Meltdown: Vulnerable Mmio stale data: Vulnerable Retbleed: Not affected Spec store bypass: Vulnerable Spectre v1: Vulnerable: __user pointer sanitization and usercopy barriers on ly; no swapgs barriers Spectre v2: Vulnerable, IBPB: disabled, STIBP: disabled, PBRSB-eIBRS: Not af fected Srbds: Not affected Tsx async abort: Vulnerable

---

uname -r 4.18.0-372.52.1.el8_6.x86_64

Best,

Wei

[zhilizheng]

Hi @wjiang-datasci,

This is a really strange problem.

Could you try my example:

wget https://plot.cnsgenomics.com/SBayesRC/example/test_tidy.ma.gz
gunzip test_tidy.ma.gz

use the test_tidy instead your summary data to run the program again.

If it still have the problem, any chance for a remote desktop sharing, we could look into the problem together under your visual supervision? Email zhili.zheng ( At ) broadinstitute.org if you think it's feasible. Or I would need your SNP list for a simulation, but I guess, it could be enviromental issue, so would like to explore this first.

Regards, Zhili

[wjiang-datasci]

Hi Zhili,

I just tried your example, still facing the same issue. I agree this is a very strange issue and should be related with environmental configuration. Since it just raises C++ exception with unknown reason, it's difficult to figure out the potential problem by myself. Remote sharing from you will be helpful. I will email you for your possible time slot. Thank you so much.

Best,

Wei

[zhilizheng]

Thanks Wei for the details. We find the issue could come from the cluster, so I close the issue.

BTW, for future readers, I find the summary data only contains 200,000 SNPs, while in HapMap3, we have over 1 Million. The missingness rate is too large for a good imputation quality (we suggest > 70% SNPs are in the summary data, or to be conservative > 80%). So, I suggest the readers to generate the LD by the genotype, which is also documented on the site (see generate LD).

Regards, Zhili