search

JonJala / mtag

Python command line tool for Multi-Trait Analysis of GWAS (MTAG)
GNU General Public License v3.0
169 stars 54 forks source link

Using MTAG to enhance the GWAS ability #213

Open ttian627 opened 3 months ago

ttian627 commented 3 months ago

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of three environments of plant hegiht which has low repeatability. We call the rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG signials for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that same as the GWAS results. My question are 1) Is MATG suitable for improving the detection ability of duplicate data？If so, Is rep_1 an abnormal result and how to handle it？

Thank you very much!

Tian

paturley commented 3 months ago

Hello Tian,

I'm a bit confused by your question. Are you worried that the very significant results for rep_1 is spurious? If you post your log file, it will help me understand what is going on better.

Best, Patrick

On Mon, May 20, 2024 at 4:06 AM real @.***> wrote:

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of three environments of plant hegiht which has low repeatability. We call the rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG signials for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that same as the GWAS results. My question are 1) Is MATG suitable for improving the detection ability of duplicate data？If so, Is rep_1 an abnormal result and how to handle it？

Thank you very much!

Tian

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5LNIGHOATZ7YCQTGRLZDGVHZAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDKMZYGM4DSMI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

ttian627 commented 3 months ago

Hello Patrick

Thanks for your reply. Yes, I think the results for rep_1 is spurious. The Manhattan plots for GWAS and MATG as well as the log file were attatched.

Thank you very much!

Best,

Tian

At 2024-05-21 02:41:50, "paturley" @.***> wrote:

Hello Tian,

I'm a bit confused by your question. Are you worried that the very significant results for rep_1 is spurious? If you post your log file, it will help me understand what is going on better.

Best, Patrick

On Mon, May 20, 2024 at 4:06 AM real @.***> wrote:

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of three environments of plant hegiht which has low repeatability. We call the rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG signials for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that same as the GWAS results. My question are 1) Is MATG suitable for improving the detection ability of duplicate data？If so, Is rep_1 an abnormal result and how to handle it？

Thank you very much!

Tian

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5LNIGHOATZ7YCQTGRLZDGVHZAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDKMZYGM4DSMI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.> 2024/05/14/03:05:46 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> <> <> MTAG: Multi-trait Analysis of GWAS <> Version: 1.0.8 <> (C) 2017 Omeed Maghzian, Raymond Walters, and Patrick Turley <> Harvard University Department of Economics / Broad Institute of MIT and Harvard <> GNU General Public License v3 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> <> Note: It is recommended to run your own QC on the input before using this program. <> Software-related correspondence: @. <> All other correspondence: @.*** <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Calling ./mtag.py \ --force \ --stream-stdout \ --ld-ref-panel /data/ttian/08-GWAS_1223/01_Emmax_1223/5M_Q_K/LDscores/ \ --sumstats phen_20_PG_tbn.txt.m.ps,phen_21_PG_tbn.txt.m.ps,phen_22_PG_tbn.txt.m.ps \ --out PG/phen_PG_tbn

phen_20_PG_tbn(rep_1),phen_21_PG_tbn(rep_2),phen_22_PG_tbn(rep_3)

2024/05/14/03:05:46 PM Beginning MTAG analysis... 2024/05/14/03:05:46 PM MTAG will use the Z column for analyses. 2024/05/14/03:05:55 PM Read in Trait 1 summary statistics (5593073 SNPs) from phen_20_PG_tbn.txt.m.ps ... 2024/05/14/03:05:55 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:05:55 PM Munging Trait 1 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:05:55 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:05:55 PM Interpreting column names as follows: 2024/05/14/03:05:55 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:05:55 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:06:04 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:06:08 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:06:09 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:06:22 PM Median value of SIGNED_SUMSTAT was 0.000275905, which seems sensible. 2024/05/14/03:06:22 PM Dropping snps with null values 2024/05/14/03:06:22 PM Metadata: 2024/05/14/03:06:23 PM Mean chi^2 = 0.941 2024/05/14/03:06:23 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:06:23 PM Lambda GC = 0.912 2024/05/14/03:06:23 PM Max chi^2 = 30.569 2024/05/14/03:06:23 PM 1 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:06:23 PM Conversion finished at Tue May 14 15:06:23 2024 2024/05/14/03:06:23 PM Total time elapsed: 28.06s 2024/05/14/03:06:33 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:33 PM Munging of Trait 1 complete. SNPs remaining: 5099629 2024/05/14/03:06:33 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Warning: The mean chi2 statistic of trait 1 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:06:51 PM Read in Trait 2 summary statistics (5593073 SNPs) from phen_21_PG_tbn.txt.m.ps ... 2024/05/14/03:06:51 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:51 PM Munging Trait 2 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:06:51 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:51 PM Interpreting column names as follows: 2024/05/14/03:06:51 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:06:51 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:07:00 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:07:03 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:07:04 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:07:18 PM Median value of SIGNED_SUMSTAT was 0.0122005, which seems sensible. 2024/05/14/03:07:18 PM Dropping snps with null values 2024/05/14/03:07:18 PM Metadata: 2024/05/14/03:07:19 PM Mean chi^2 = 0.933 2024/05/14/03:07:19 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:07:19 PM Lambda GC = 0.91 2024/05/14/03:07:19 PM Max chi^2 = 25.301 2024/05/14/03:07:19 PM 0 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:07:19 PM Conversion finished at Tue May 14 15:07:19 2024 2024/05/14/03:07:19 PM Total time elapsed: 28.61s 2024/05/14/03:07:28 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:28 PM Munging of Trait 2 complete. SNPs remaining: 5099629 2024/05/14/03:07:28 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Warning: The mean chi2 statistic of trait 2 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:07:47 PM Read in Trait 3 summary statistics (5593073 SNPs) from phen_22_PG_tbn.txt.m.ps ... 2024/05/14/03:07:47 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:47 PM Munging Trait 3 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:07:47 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:47 PM Interpreting column names as follows: 2024/05/14/03:07:47 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:07:47 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:07:58 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:08:02 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:08:03 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:08:16 PM Median value of SIGNED_SUMSTAT was 0.00952361, which seems sensible. 2024/05/14/03:08:17 PM Dropping snps with null values 2024/05/14/03:08:17 PM Metadata: 2024/05/14/03:08:18 PM Mean chi^2 = 0.951 2024/05/14/03:08:18 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:08:18 PM Lambda GC = 0.934 2024/05/14/03:08:18 PM Max chi^2 = 31.856 2024/05/14/03:08:18 PM 3 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:08:18 PM Conversion finished at Tue May 14 15:08:18 2024 2024/05/14/03:08:18 PM Total time elapsed: 31.07s 2024/05/14/03:08:29 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:08:29 PM Munging of Trait 3 complete. SNPs remaining: 5099629 2024/05/14/03:08:29 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:30 PM Warning: The mean chi2 statistic of trait 3 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:08:46 PM Dropped 737230 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait1 2024/05/14/03:09:03 PM Dropped 0 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait2 2024/05/14/03:09:22 PM Dropped 0 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait3 2024/05/14/03:09:22 PM ... Merge of GWAS summary statistics complete. Number of SNPs: 4362399 2024/05/14/03:09:38 PM Using 4362399 SNPs to estimate Omega (0 SNPs excluded due to strand ambiguity) 2024/05/14/03:09:38 PM Estimating sigma.. 2024/05/14/03:14:29 PM Checking for positive definiteness .. 2024/05/14/03:14:29 PM Sigma hat: [[1.165 0.97 0.88 ] [0.97 1.177 0.914] [0.88 0.914 1.131]] 2024/05/14/03:14:29 PM Mean chi^2 of SNPs used to estimate Omega is low for some SNPsMTAG may not perform well in this situation. 2024/05/14/03:14:30 PM Beginning estimation of Omega ... 2024/05/14/03:14:30 PM Using GMM estimator of Omega .. 2024/05/14/03:14:31 PM Checking for positive definiteness .. 2024/05/14/03:14:31 PM matrix is not positive definite, performing adjustment.. 2024/05/14/03:14:31 PM Warning: max number of iterations reached in adjustment procedure. Sigma matrix used is still non-positive-definite. 2024/05/14/03:14:31 PM Completed estimation of Omega ... 2024/05/14/03:14:31 PM Beginning MTAG calculations... 2024/05/14/03:15:19 PM ... Completed MTAG calculations. 2024/05/14/03:15:19 PM Writing Phenotype 1 to file ... 2024/05/14/03:15:58 PM Writing Phenotype 2 to file ... 2024/05/14/03:16:35 PM Writing Phenotype 3 to file ... 2024/05/14/03:17:15 PM Summary of MTAG results:

Trait # SNPs used N (max) N (mean) GWAS mean chi^2 MTAG mean chi^2 GWAS equiv. (max) N 1 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.807 18.926 -113599
2 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.792 1.771 -4537
3 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.841 1.444 -3409

Estimated Omega: [[-1.839e-04 -6.803e-09 -5.303e-09] [-6.803e-09 -2.001e-04 -5.786e-09] [-5.303e-09 -5.786e-09 -1.472e-04]]

(Correlation): [[nan nan nan] [nan nan nan] [nan nan nan]]

Estimated Sigma: [[1.165 0.97 0.88 ] [0.97 1.177 0.914] [0.88 0.914 1.131]]

(Correlation): [[1. 0.828 0.767] [0.828 1. 0.792] [0.767 0.792 1. ]]

MTAG weight factors: (average across SNPs) [-0.014 -0.035 0.061]

2024/05/14/03:17:15 PM
2024/05/14/03:17:15 PM MTAG results saved to file. 2024/05/14/03:17:15 PM MTAG complete. Time elapsed: 11.0m:29.370085001s

paturley commented 3 months ago

Ah. So the problem is that your mean chi2 statistics are less than one, so the implicit heritability is negative. I generally recommend that MTAG not be used unless the mean chi2 is at least 1.1 for each trait. Better if it's even higher than that. In your particular case, I think MTAG may not be the right tool.

On Mon, May 20, 2024 at 9:53 PM real @.***> wrote:

Hello Patrick

Thanks for your reply. Yes, I think the results for rep_1 is spurious. The Manhattan plots for GWAS and MATG as well as the log file were attatched.

Thank you very much!

Best,

Tian

At 2024-05-21 02:41:50, "paturley" @.***> wrote:

Hello Tian,

I'm a bit confused by your question. Are you worried that the very significant results for rep_1 is spurious? If you post your log file, it will help me understand what is going on better.

Best, Patrick

On Mon, May 20, 2024 at 4:06 AM real @.***> wrote:

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of three environments of plant hegiht which has low repeatability. We call the rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG signials for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that same as the GWAS results. My question are 1) Is MATG suitable for improving the detection ability of duplicate data？If so, Is rep_1 an abnormal result and how to handle it？

Thank you very much!

Tian

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AFBUB5LNIGHOATZ7YCQTGRLZDGVHZAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDKMZYGM4DSMI

. You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***> 2024/05/14/03:05:46 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> <> <> MTAG: Multi-trait Analysis of GWAS <> Version: 1.0.8 <> (C) 2017 Omeed Maghzian, Raymond Walters, and Patrick Turley <> Harvard University Department of Economics / Broad Institute of MIT and Harvard <> GNU General Public License v3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> <> Note: It is recommended to run your own QC on the input before using this program. <> Software-related correspondence: @. <> All other correspondence: @.

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Calling ./mtag.py \ --force \ --stream-stdout \ --ld-ref-panel /data/ttian/08-GWAS_1223/01_Emmax_1223/5M_Q_K/LDscores/ \ --sumstats phen_20_PG_tbn.txt.m.ps,phen_21_PG_tbn.txt.m.ps, phen_22_PG_tbn.txt.m.ps \ --out PG/phen_PG_tbn

phen_20_PG_tbn(rep_1),phen_21_PG_tbn(rep_2),phen_22_PG_tbn(rep_3)

2024/05/14/03:05:46 PM Beginning MTAG analysis... 2024/05/14/03:05:46 PM MTAG will use the Z column for analyses. 2024/05/14/03:05:55 PM Read in Trait 1 summary statistics (5593073 SNPs) from phen_20_PG_tbn.txt.m.ps ... 2024/05/14/03:05:55 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:05:55 PM Munging Trait 1 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:05:55 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:05:55 PM Interpreting column names as follows: 2024/05/14/03:05:55 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:05:55 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:06:04 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:06:08 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:06:09 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:06:22 PM Median value of SIGNED_SUMSTAT was 0.000275905, which seems sensible. 2024/05/14/03:06:22 PM Dropping snps with null values 2024/05/14/03:06:22 PM Metadata: 2024/05/14/03:06:23 PM Mean chi^2 = 0.941 2024/05/14/03:06:23 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:06:23 PM Lambda GC = 0.912 2024/05/14/03:06:23 PM Max chi^2 = 30.569 2024/05/14/03:06:23 PM 1 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:06:23 PM Conversion finished at Tue May 14 15:06:23 2024 2024/05/14/03:06:23 PM Total time elapsed: 28.06s 2024/05/14/03:06:33 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:33 PM Munging of Trait 1 complete. SNPs remaining: 5099629 2024/05/14/03:06:33 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Warning: The mean chi2 statistic of trait 1 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:06:51 PM Read in Trait 2 summary statistics (5593073 SNPs) from phen_21_PG_tbn.txt.m.ps ... 2024/05/14/03:06:51 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:51 PM Munging Trait 2 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:06:51 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:51 PM Interpreting column names as follows: 2024/05/14/03:06:51 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:06:51 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:07:00 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:07:03 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:07:04 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:07:18 PM Median value of SIGNED_SUMSTAT was 0.0122005, which seems sensible. 2024/05/14/03:07:18 PM Dropping snps with null values 2024/05/14/03:07:18 PM Metadata: 2024/05/14/03:07:19 PM Mean chi^2 = 0.933 2024/05/14/03:07:19 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:07:19 PM Lambda GC = 0.91 2024/05/14/03:07:19 PM Max chi^2 = 25.301 2024/05/14/03:07:19 PM 0 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:07:19 PM Conversion finished at Tue May 14 15:07:19 2024 2024/05/14/03:07:19 PM Total time elapsed: 28.61s 2024/05/14/03:07:28 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:28 PM Munging of Trait 2 complete. SNPs remaining: 5099629 2024/05/14/03:07:28 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Warning: The mean chi2 statistic of trait 2 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:07:47 PM Read in Trait 3 summary statistics (5593073 SNPs) from phen_22_PG_tbn.txt.m.ps ... 2024/05/14/03:07:47 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:47 PM Munging Trait 3 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:07:47 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:47 PM Interpreting column names as follows: 2024/05/14/03:07:47 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:07:47 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:07:58 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:08:02 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:08:03 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:08:16 PM Median value of SIGNED_SUMSTAT was 0.00952361, which seems sensible. 2024/05/14/03:08:17 PM Dropping snps with null values 2024/05/14/03:08:17 PM Metadata: 2024/05/14/03:08:18 PM Mean chi^2 = 0.951 2024/05/14/03:08:18 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:08:18 PM Lambda GC = 0.934 2024/05/14/03:08:18 PM Max chi^2 = 31.856 2024/05/14/03:08:18 PM 3 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:08:18 PM Conversion finished at Tue May 14 15:08:18 2024 2024/05/14/03:08:18 PM Total time elapsed: 31.07s 2024/05/14/03:08:29 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:08:29 PM Munging of Trait 3 complete. SNPs remaining: 5099629 2024/05/14/03:08:29 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:30 PM Warning: The mean chi2 statistic of trait 3 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:08:46 PM Dropped 737230 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait1 2024/05/14/03:09:03 PM Dropped 0 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait2 2024/05/14/03:09:22 PM Dropped 0 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait3 2024/05/14/03:09:22 PM ... Merge of GWAS summary statistics complete. Number of SNPs: 4362399 2024/05/14/03:09:38 PM Using 4362399 SNPs to estimate Omega (0 SNPs excluded due to strand ambiguity) 2024/05/14/03:09:38 PM Estimating sigma.. 2024/05/14/03:14:29 PM Checking for positive definiteness .. 2024/05/14/03:14:29 PM Sigma hat: [[1.165 0.97 0.88 ] [0.97 1.177 0.914] [0.88 0.914 1.131]] 2024/05/14/03:14:29 PM Mean chi^2 of SNPs used to estimate Omega is low for some SNPsMTAG may not perform well in this situation. 2024/05/14/03:14:30 PM Beginning estimation of Omega ... 2024/05/14/03:14:30 PM Using GMM estimator of Omega .. 2024/05/14/03:14:31 PM Checking for positive definiteness .. 2024/05/14/03:14:31 PM matrix is not positive definite, performing adjustment.. 2024/05/14/03:14:31 PM Warning: max number of iterations reached in adjustment procedure. Sigma matrix used is still non-positive-definite. 2024/05/14/03:14:31 PM Completed estimation of Omega ... 2024/05/14/03:14:31 PM Beginning MTAG calculations... 2024/05/14/03:15:19 PM ... Completed MTAG calculations. 2024/05/14/03:15:19 PM Writing Phenotype 1 to file ... 2024/05/14/03:15:58 PM Writing Phenotype 2 to file ... 2024/05/14/03:16:35 PM Writing Phenotype 3 to file ... 2024/05/14/03:17:15 PM Summary of MTAG results:

Trait # SNPs used N (max) N (mean) GWAS mean chi^2 MTAG mean chi^2 GWAS equiv. (max) N 1 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.807 18.926 -113599 2 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.792 1.771 -4537 3 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.841 1.444 -3409

Estimated Omega: [[-1.839e-04 -6.803e-09 -5.303e-09] [-6.803e-09 -2.001e-04 -5.786e-09] [-5.303e-09 -5.786e-09 -1.472e-04]]

(Correlation): [[nan nan nan] [nan nan nan] [nan nan nan]]

Estimated Sigma: [[1.165 0.97 0.88 ] [0.97 1.177 0.914] [0.88 0.914 1.131]]

(Correlation): [[1. 0.828 0.767] [0.828 1. 0.792] [0.767 0.792 1. ]]

MTAG weight factors: (average across SNPs) [-0.014 -0.035 0.061]

2024/05/14/03:17:15 PM 2024/05/14/03:17:15 PM MTAG results saved to file. 2024/05/14/03:17:15 PM MTAG complete. Time elapsed: 11.0m:29.370085001s

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213#issuecomment-2121555944, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5KI65WG77QJPAX3243ZDKSJDAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRRGU2TKOJUGQ . You are receiving this because you commented.Message ID: @.***>

ttian627 commented 3 months ago

Hello Patrick

Thanks for your reply. I find that the mean chr^2 of my GWAS results for all the traits were less than 1. My phenotype data includes plant height, leaf angle and tassel branch numbers, etc which have higher broad sense heritability.Where could this be the problems?

Thank you and all the best,

Tian

At 2024-05-22 04:40:51, "paturley" @.***> wrote:

Ah. So the problem is that your mean chi2 statistics are less than one, so the implicit heritability is negative. I generally recommend that MTAG not be used unless the mean chi2 is at least 1.1 for each trait. Better if it's even higher than that. In your particular case, I think MTAG may not be the right tool.

On Mon, May 20, 2024 at 9:53 PM real @.***> wrote:

Hello Patrick

Thanks for your reply. Yes, I think the results for rep_1 is spurious. The Manhattan plots for GWAS and MATG as well as the log file were attatched.

Thank you very much!

Best,

Tian

At 2024-05-21 02:41:50, "paturley" @.***> wrote:

Hello Tian,

I'm a bit confused by your question. Are you worried that the very significant results for rep_1 is spurious? If you post your log file, it will help me understand what is going on better.

Best, Patrick

On Mon, May 20, 2024 at 4:06 AM real @.***> wrote:

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of three environments of plant hegiht which has low repeatability. We call the rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG signials for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that same as the GWAS results. My question are 1) Is MATG suitable for improving the detection ability of duplicate data？If so, Is rep_1 an abnormal result and how to handle it？

Thank you very much!

Tian

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AFBUB5LNIGHOATZ7YCQTGRLZDGVHZAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDKMZYGM4DSMI

. You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***> 2024/05/14/03:05:46 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> <> <> MTAG: Multi-trait Analysis of GWAS <> Version: 1.0.8 <> (C) 2017 Omeed Maghzian, Raymond Walters, and Patrick Turley <> Harvard University Department of Economics / Broad Institute of MIT and Harvard <> GNU General Public License v3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> <> Note: It is recommended to run your own QC on the input before using this program. <> Software-related correspondence: @. <> All other correspondence: @.

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Calling ./mtag.py \ --force \ --stream-stdout \ --ld-ref-panel /data/ttian/08-GWAS_1223/01_Emmax_1223/5M_Q_K/LDscores/ \ --sumstats phen_20_PG_tbn.txt.m.ps,phen_21_PG_tbn.txt.m.ps, phen_22_PG_tbn.txt.m.ps \ --out PG/phen_PG_tbn

phen_20_PG_tbn(rep_1),phen_21_PG_tbn(rep_2),phen_22_PG_tbn(rep_3)

2024/05/14/03:05:46 PM Beginning MTAG analysis... 2024/05/14/03:05:46 PM MTAG will use the Z column for analyses. 2024/05/14/03:05:55 PM Read in Trait 1 summary statistics (5593073 SNPs) from phen_20_PG_tbn.txt.m.ps ... 2024/05/14/03:05:55 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:05:55 PM Munging Trait 1 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:05:55 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:05:55 PM Interpreting column names as follows: 2024/05/14/03:05:55 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:05:55 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:06:04 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:06:08 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:06:09 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:06:22 PM Median value of SIGNED_SUMSTAT was 0.000275905, which seems sensible. 2024/05/14/03:06:22 PM Dropping snps with null values 2024/05/14/03:06:22 PM Metadata: 2024/05/14/03:06:23 PM Mean chi^2 = 0.941 2024/05/14/03:06:23 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:06:23 PM Lambda GC = 0.912 2024/05/14/03:06:23 PM Max chi^2 = 30.569 2024/05/14/03:06:23 PM 1 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:06:23 PM Conversion finished at Tue May 14 15:06:23 2024 2024/05/14/03:06:23 PM Total time elapsed: 28.06s 2024/05/14/03:06:33 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:33 PM Munging of Trait 1 complete. SNPs remaining: 5099629 2024/05/14/03:06:33 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Warning: The mean chi2 statistic of trait 1 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:06:51 PM Read in Trait 2 summary statistics (5593073 SNPs) from phen_21_PG_tbn.txt.m.ps ... 2024/05/14/03:06:51 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:51 PM Munging Trait 2 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:06:51 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:51 PM Interpreting column names as follows: 2024/05/14/03:06:51 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:06:51 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:07:00 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:07:03 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:07:04 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:07:18 PM Median value of SIGNED_SUMSTAT was 0.0122005, which seems sensible. 2024/05/14/03:07:18 PM Dropping snps with null values 2024/05/14/03:07:18 PM Metadata: 2024/05/14/03:07:19 PM Mean chi^2 = 0.933 2024/05/14/03:07:19 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:07:19 PM Lambda GC = 0.91 2024/05/14/03:07:19 PM Max chi^2 = 25.301 2024/05/14/03:07:19 PM 0 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:07:19 PM Conversion finished at Tue May 14 15:07:19 2024 2024/05/14/03:07:19 PM Total time elapsed: 28.61s 2024/05/14/03:07:28 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:28 PM Munging of Trait 2 complete. SNPs remaining: 5099629 2024/05/14/03:07:28 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Warning: The mean chi2 statistic of trait 2 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:07:47 PM Read in Trait 3 summary statistics (5593073 SNPs) from phen_22_PG_tbn.txt.m.ps ... 2024/05/14/03:07:47 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:47 PM Munging Trait 3 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:07:47 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:47 PM Interpreting column names as follows: 2024/05/14/03:07:47 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:07:47 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:07:58 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:08:02 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:08:03 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:08:16 PM Median value of SIGNED_SUMSTAT was 0.00952361, which seems sensible. 2024/05/14/03:08:17 PM Dropping snps with null values 2024/05/14/03:08:17 PM Metadata: 2024/05/14/03:08:18 PM Mean chi^2 = 0.951 2024/05/14/03:08:18 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:08:18 PM Lambda GC = 0.934 2024/05/14/03:08:18 PM Max chi^2 = 31.856 2024/05/14/03:08:18 PM 3 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:08:18 PM Conversion finished at Tue May 14 15:08:18 2024 2024/05/14/03:08:18 PM Total time elapsed: 31.07s 2024/05/14/03:08:29 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:08:29 PM Munging of Trait 3 complete. SNPs remaining: 5099629 2024/05/14/03:08:29 PM <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:30 PM Warning: The mean chi2 statistic of trait 3 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:08:46 PM Dropped 737230 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait1 2024/05/14/03:09:03 PM Dropped 0 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait2 2024/05/14/03:09:22 PM Dropped 0 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait3 2024/05/14/03:09:22 PM ... Merge of GWAS summary statistics complete. Number of SNPs: 4362399 2024/05/14/03:09:38 PM Using 4362399 SNPs to estimate Omega (0 SNPs excluded due to strand ambiguity) 2024/05/14/03:09:38 PM Estimating sigma.. 2024/05/14/03:14:29 PM Checking for positive definiteness .. 2024/05/14/03:14:29 PM Sigma hat: [[1.165 0.97 0.88 ] [0.97 1.177 0.914] [0.88 0.914 1.131]] 2024/05/14/03:14:29 PM Mean chi^2 of SNPs used to estimate Omega is low for some SNPsMTAG may not perform well in this situation. 2024/05/14/03:14:30 PM Beginning estimation of Omega ... 2024/05/14/03:14:30 PM Using GMM estimator of Omega .. 2024/05/14/03:14:31 PM Checking for positive definiteness .. 2024/05/14/03:14:31 PM matrix is not positive definite, performing adjustment.. 2024/05/14/03:14:31 PM Warning: max number of iterations reached in adjustment procedure. Sigma matrix used is still non-positive-definite. 2024/05/14/03:14:31 PM Completed estimation of Omega ... 2024/05/14/03:14:31 PM Beginning MTAG calculations... 2024/05/14/03:15:19 PM ... Completed MTAG calculations. 2024/05/14/03:15:19 PM Writing Phenotype 1 to file ... 2024/05/14/03:15:58 PM Writing Phenotype 2 to file ... 2024/05/14/03:16:35 PM Writing Phenotype 3 to file ... 2024/05/14/03:17:15 PM Summary of MTAG results:

Trait # SNPs used N (max) N (mean) GWAS mean chi^2 MTAG mean chi^2 GWAS equiv. (max) N 1 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.807 18.926 -113599 2 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.792 1.771 -4537 3 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.841 1.444 -3409

Estimated Omega: [[-1.839e-04 -6.803e-09 -5.303e-09] [-6.803e-09 -2.001e-04 -5.786e-09] [-5.303e-09 -5.786e-09 -1.472e-04]]

(Correlation): [[nan nan nan] [nan nan nan] [nan nan nan]]

Estimated Sigma: [[1.165 0.97 0.88 ] [0.97 1.177 0.914] [0.88 0.914 1.131]]

(Correlation): [[1. 0.828 0.767] [0.828 1. 0.792] [0.767 0.792 1. ]]

MTAG weight factors: (average across SNPs) [-0.014 -0.035 0.061]

2024/05/14/03:17:15 PM 2024/05/14/03:17:15 PM MTAG results saved to file. 2024/05/14/03:17:15 PM MTAG complete. Time elapsed: 11.0m:29.370085001s

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213#issuecomment-2121555944, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5KI65WG77QJPAX3243ZDKSJDAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRRGU2TKOJUGQ . You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

paturley commented 3 months ago

Mean chi2 is both a function of heritability and sample size. Your sample size must be quite small then.

On Wed, May 22, 2024, 5:27 AM real @.***> wrote:

Hello Patrick

Thanks for your reply. I find that the mean chr^2 of my GWAS results for all the traits were less than 1. My phenotype data includes plant height, leaf angle and tassel branch numbers, etc which have higher broad sense heritability.Where could this be the problems?

Thank you and all the best,

Tian

At 2024-05-22 04:40:51, "paturley" @.***> wrote:

Ah. So the problem is that your mean chi2 statistics are less than one, so the implicit heritability is negative. I generally recommend that MTAG not be used unless the mean chi2 is at least 1.1 for each trait. Better if it's even higher than that. In your particular case, I think MTAG may not be the right tool.

On Mon, May 20, 2024 at 9:53 PM real @.***> wrote:

Hello Patrick

Thanks for your reply. Yes, I think the results for rep_1 is spurious. The Manhattan plots for GWAS and MATG as well as the log file were attatched.

Thank you very much!

Best,

Tian

At 2024-05-21 02:41:50, "paturley" @.***> wrote:

Hello Tian,

I'm a bit confused by your question. Are you worried that the very significant results for rep_1 is spurious? If you post your log file, it will help me understand what is going on better.

Best, Patrick

On Mon, May 20, 2024 at 4:06 AM real @.***> wrote:

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of three environments of plant hegiht which has low repeatability. We call the rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG signials for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that same as the GWAS results. My question are 1) Is MATG suitable for improving the detection ability of duplicate data？If so, Is rep_1 an abnormal result and how to handle it？

Thank you very much!

Tian

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213, or unsubscribe <

https://github.com/notifications/unsubscribe-auth/AFBUB5LNIGHOATZ7YCQTGRLZDGVHZAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDKMZYGM4DSMI

. You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***> 2024/05/14/03:05:46 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> <> <> MTAG: Multi-trait Analysis of GWAS <> Version: 1.0.8 <> (C) 2017 Omeed Maghzian, Raymond Walters, and Patrick Turley <> Harvard University Department of Economics / Broad Institute of MIT and Harvard <> GNU General Public License v3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> <> Note: It is recommended to run your own QC on the input before using this program. <> Software-related correspondence: @. <> All other correspondence: @.

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Calling ./mtag.py \ --force \ --stream-stdout \ --ld-ref-panel /data/ttian/08-GWAS_1223/01_Emmax_1223/5M_Q_K/LDscores/ \ --sumstats phen_20_PG_tbn.txt.m.ps,phen_21_PG_tbn.txt.m.ps, phen_22_PG_tbn.txt.m.ps \ --out PG/phen_PG_tbn

phen_20_PG_tbn(rep_1),phen_21_PG_tbn(rep_2),phen_22_PG_tbn(rep_3)

2024/05/14/03:05:46 PM Beginning MTAG analysis... 2024/05/14/03:05:46 PM MTAG will use the Z column for analyses. 2024/05/14/03:05:55 PM Read in Trait 1 summary statistics (5593073 SNPs) from phen_20_PG_tbn.txt.m.ps ... 2024/05/14/03:05:55 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:05:55 PM Munging Trait 1 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:05:55 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:05:55 PM Interpreting column names as follows: 2024/05/14/03:05:55 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:05:55 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:06:04 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:06:08 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:06:09 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:06:22 PM Median value of SIGNED_SUMSTAT was 0.000275905, which seems sensible. 2024/05/14/03:06:22 PM Dropping snps with null values 2024/05/14/03:06:22 PM Metadata: 2024/05/14/03:06:23 PM Mean chi^2 = 0.941 2024/05/14/03:06:23 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:06:23 PM Lambda GC = 0.912 2024/05/14/03:06:23 PM Max chi^2 = 30.569 2024/05/14/03:06:23 PM 1 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:06:23 PM Conversion finished at Tue May 14 15:06:23 2024 2024/05/14/03:06:23 PM Total time elapsed: 28.06s 2024/05/14/03:06:33 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:33 PM Munging of Trait 1 complete. SNPs remaining: 5099629 2024/05/14/03:06:33 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Warning: The mean chi2 statistic of trait 1 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:06:51 PM Read in Trait 2 summary statistics (5593073 SNPs) from phen_21_PG_tbn.txt.m.ps ... 2024/05/14/03:06:51 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:51 PM Munging Trait 2 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:06:51 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:06:51 PM Interpreting column names as follows: 2024/05/14/03:06:51 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:06:51 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:07:00 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:07:03 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:07:04 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:07:18 PM Median value of SIGNED_SUMSTAT was 0.0122005, which seems sensible. 2024/05/14/03:07:18 PM Dropping snps with null values 2024/05/14/03:07:18 PM Metadata: 2024/05/14/03:07:19 PM Mean chi^2 = 0.933 2024/05/14/03:07:19 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:07:19 PM Lambda GC = 0.91 2024/05/14/03:07:19 PM Max chi^2 = 25.301 2024/05/14/03:07:19 PM 0 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:07:19 PM Conversion finished at Tue May 14 15:07:19 2024 2024/05/14/03:07:19 PM Total time elapsed: 28.61s 2024/05/14/03:07:28 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:28 PM Munging of Trait 2 complete. SNPs remaining: 5099629 2024/05/14/03:07:28 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Warning: The mean chi2 statistic of trait 2 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:07:47 PM Read in Trait 3 summary statistics (5593073 SNPs) from phen_22_PG_tbn.txt.m.ps ... 2024/05/14/03:07:47 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:47 PM Munging Trait 3 <><><<>><><><><><><><><><><><><><><><><><><><><><><><><><>< 2024/05/14/03:07:47 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:07:47 PM Interpreting column names as follows: 2024/05/14/03:07:47 PM snpid: Variant ID (e.g., rs number) n: Sample size a1: a1, interpreted as ref allele for signed sumstat. pval: p-Value a2: a2, interpreted as non-ref allele for signed sumstat. z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:07:47 PM Reading sumstats from provided DataFrame into memory 10000000 SNPs at a time. 2024/05/14/03:07:58 PM Read 5593073 SNPs from --sumstats file. Removed 0 SNPs with missing values. Removed 0 SNPs with INFO <= None. Removed 0 SNPs with MAF <= 0.01. Removed 0 SNPs with SE <0 or NaN values. Removed 0 SNPs with out-of-bounds p-values. Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs were not dropped. 5099629 SNPs remain. 2024/05/14/03:08:02 PM Removed 0 SNPs with duplicated rs numbers (5099629 SNPs remain). 2024/05/14/03:08:03 PM Removed 0 SNPs with N < 815.333333333 (5099629 SNPs remain). 2024/05/14/03:08:16 PM Median value of SIGNED_SUMSTAT was 0.00952361, which seems sensible. 2024/05/14/03:08:17 PM Dropping snps with null values 2024/05/14/03:08:17 PM Metadata: 2024/05/14/03:08:18 PM Mean chi^2 = 0.951 2024/05/14/03:08:18 PM WARNING: mean chi^2 may be too small. 2024/05/14/03:08:18 PM Lambda GC = 0.934 2024/05/14/03:08:18 PM Max chi^2 = 31.856 2024/05/14/03:08:18 PM 3 Genome-wide significant SNPs (some may have been removed by filtering). 2024/05/14/03:08:18 PM Conversion finished at Tue May 14 15:08:18 2024 2024/05/14/03:08:18 PM Total time elapsed: 31.07s 2024/05/14/03:08:29 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> 2024/05/14/03:08:29 PM Munging of Trait 3 complete. SNPs remaining: 5099629 2024/05/14/03:08:29 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:30 PM Warning: The mean chi2 statistic of trait 3 is less 1.02 - MTAG estimates may be unstable. 2024/05/14/03:08:46 PM Dropped 737230 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait1 2024/05/14/03:09:03 PM Dropped 0 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait2 2024/05/14/03:09:22 PM Dropped 0 SNPs due to strand ambiguity, 4362399 SNPs remain in intersection after merging trait3 2024/05/14/03:09:22 PM ... Merge of GWAS summary statistics complete. Number of SNPs: 4362399 2024/05/14/03:09:38 PM Using 4362399 SNPs to estimate Omega (0 SNPs excluded due to strand ambiguity) 2024/05/14/03:09:38 PM Estimating sigma.. 2024/05/14/03:14:29 PM Checking for positive definiteness .. 2024/05/14/03:14:29 PM Sigma hat: [[1.165 0.97 0.88 ] [0.97 1.177 0.914] [0.88 0.914 1.131]] 2024/05/14/03:14:29 PM Mean chi^2 of SNPs used to estimate Omega is low for some SNPsMTAG may not perform well in this situation. 2024/05/14/03:14:30 PM Beginning estimation of Omega ... 2024/05/14/03:14:30 PM Using GMM estimator of Omega .. 2024/05/14/03:14:31 PM Checking for positive definiteness .. 2024/05/14/03:14:31 PM matrix is not positive definite, performing adjustment.. 2024/05/14/03:14:31 PM Warning: max number of iterations reached in adjustment procedure. Sigma matrix used is still non-positive-definite. 2024/05/14/03:14:31 PM Completed estimation of Omega ... 2024/05/14/03:14:31 PM Beginning MTAG calculations... 2024/05/14/03:15:19 PM ... Completed MTAG calculations. 2024/05/14/03:15:19 PM Writing Phenotype 1 to file ... 2024/05/14/03:15:58 PM Writing Phenotype 2 to file ... 2024/05/14/03:16:35 PM Writing Phenotype 3 to file ... 2024/05/14/03:17:15 PM Summary of MTAG results:

Trait # SNPs used N (max) N (mean) GWAS mean chi^2 MTAG mean chi^2 GWAS equiv. (max) N 1 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.807 18.926 -113599 2 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.792 1.771 -4537 3 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.841 1.444 -3409

Estimated Omega: [[-1.839e-04 -6.803e-09 -5.303e-09] [-6.803e-09 -2.001e-04 -5.786e-09] [-5.303e-09 -5.786e-09 -1.472e-04]]

(Correlation): [[nan nan nan] [nan nan nan] [nan nan nan]]

Estimated Sigma: [[1.165 0.97 0.88 ] [0.97 1.177 0.914] [0.88 0.914 1.131]]

(Correlation): [[1. 0.828 0.767] [0.828 1. 0.792] [0.767 0.792 1. ]]

MTAG weight factors: (average across SNPs) [-0.014 -0.035 0.061]

2024/05/14/03:17:15 PM 2024/05/14/03:17:15 PM MTAG results saved to file. 2024/05/14/03:17:15 PM MTAG complete. Time elapsed: 11.0m:29.370085001s

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213#issuecomment-2121555944, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AFBUB5KI65WG77QJPAX3243ZDKSJDAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRRGU2TKOJUGQ

. You are receiving this because you commented.Message ID: @.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213#issuecomment-2124303541, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5OVPXTWXFT7VEGB37LZDRQG3AVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRUGMYDGNJUGE . You are receiving this because you commented.Message ID: @.***>

ttian627 commented 3 months ago

I have more than 1000 samples which are relatively large for maize population.Is there other problems？Thank youTian在 2024年5月22日，18:49，paturley @.***> 写道： Mean chi2 is both a function of heritability and sample size. Your sample

size must be quite small then.

On Wed, May 22, 2024, 5:27 AM real @.***> wrote:

Hello Patrick

Thanks for your reply. I find that the mean chr^2 of my GWAS results for

all the traits were less than 1. My phenotype data includes plant height,

leaf angle and tassel branch numbers, etc which have higher broad sense

heritability.Where could this be the problems?

Thank you and all the best,

Tian

At 2024-05-22 04:40:51, "paturley" @.***> wrote:

Ah. So the problem is that your mean chi2 statistics are less than one, so

the implicit heritability is negative. I generally recommend that MTAG not

be used unless the mean chi2 is at least 1.1 for each trait. Better if it's

even higher than that. In your particular case, I think MTAG may not be the

right tool.

On Mon, May 20, 2024 at 9:53 PM real @.***> wrote:

Hello Patrick

Thanks for your reply. Yes, I think the results for rep_1 is spurious.

The

Manhattan plots for GWAS and MATG as well as the log file were attatched.

Thank you very much!

Best,

Tian

At 2024-05-21 02:41:50, "paturley" @.***> wrote:

Hello Tian,

I'm a bit confused by your question. Are you worried that the very

significant results for rep_1 is spurious? If you post your log file, it

will help me understand what is going on better.

Best,

Patrick

On Mon, May 20, 2024 at 4:06 AM real @.***> wrote:

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of

three

environments of plant hegiht which has low repeatability. We call the

rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG

signials

for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that same

as

the GWAS results. My question are 1) Is MATG suitable for improving the

detection ability of duplicate data？If so, Is rep_1 an abnormal result

and

how to handle it？

Thank you very much!

Tian

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213, or unsubscribe

<

https://github.com/notifications/unsubscribe-auth/AFBUB5LNIGHOATZ7YCQTGRLZDGVHZAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDKMZYGM4DSMI

.

You are receiving this because you are subscribed to this

thread.Message

ID: @.***>

—

Reply to this email directly, view it on GitHub, or unsubscribe.

You are receiving this because you authored the thread.Message ID:

@.***>

2024/05/14/03:05:46 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

<>

<> MTAG: Multi-trait Analysis of GWAS

<> Version: 1.0.8

<> (C) 2017 Omeed Maghzian, Raymond Walters, and Patrick Turley

<> Harvard University Department of Economics / Broad Institute of MIT

and

Harvard

<> GNU General Public License v3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

<> Note: It is recommended to run your own QC on the input before using

this program.

<> Software-related correspondence: @.***

<> All other correspondence: @.***

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Calling ./mtag.py \

--force \

--stream-stdout \

--ld-ref-panel /data/ttian/08-GWAS_1223/01_Emmax_1223/5M_Q_K/LDscores/ \

--sumstats phen_20_PG_tbn.txt.m.ps,phen_21_PG_tbn.txt.m.ps,

phen_22_PG_tbn.txt.m.ps \

--out PG/phen_PG_tbn

phen_20_PG_tbn(rep_1),phen_21_PG_tbn(rep_2),phen_22_PG_tbn(rep_3)

2024/05/14/03:05:46 PM Beginning MTAG analysis...

2024/05/14/03:05:46 PM MTAG will use the Z column for analyses.

2024/05/14/03:05:55 PM Read in Trait 1 summary statistics (5593073 SNPs)

from phen_20_PG_tbn.txt.m.ps ...

2024/05/14/03:05:55 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:05:55 PM Munging Trait 1

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:05:55 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:05:55 PM Interpreting column names as follows:

2024/05/14/03:05:55 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:05:55 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:06:04 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:06:08 PM Removed 0 SNPs with duplicated rs numbers (5099629

SNPs remain).

2024/05/14/03:06:09 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:06:22 PM Median value of SIGNED_SUMSTAT was 0.000275905,

which seems sensible.

2024/05/14/03:06:22 PM Dropping snps with null values

2024/05/14/03:06:22 PM

Metadata:

2024/05/14/03:06:23 PM Mean chi^2 = 0.941

2024/05/14/03:06:23 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:06:23 PM Lambda GC = 0.912

2024/05/14/03:06:23 PM Max chi^2 = 30.569

2024/05/14/03:06:23 PM 1 Genome-wide significant SNPs (some may have been

removed by filtering).

2024/05/14/03:06:23 PM

Conversion finished at Tue May 14 15:06:23 2024

2024/05/14/03:06:23 PM Total time elapsed: 28.06s

2024/05/14/03:06:33 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Munging of Trait 1 complete. SNPs remaining:

5099629

2024/05/14/03:06:33 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Warning: The mean chi2 statistic of trait 1 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:06:51 PM Read in Trait 2 summary statistics (5593073 SNPs)

from phen_21_PG_tbn.txt.m.ps ...

2024/05/14/03:06:51 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:51 PM Munging Trait 2

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:06:51 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:51 PM Interpreting column names as follows:

2024/05/14/03:06:51 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:06:51 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:07:00 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:07:03 PM Removed 0 SNPs with duplicated rs numbers (5099629

SNPs remain).

2024/05/14/03:07:04 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:07:18 PM Median value of SIGNED_SUMSTAT was 0.0122005,

which

seems sensible.

2024/05/14/03:07:18 PM Dropping snps with null values

2024/05/14/03:07:18 PM

Metadata:

2024/05/14/03:07:19 PM Mean chi^2 = 0.933

2024/05/14/03:07:19 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:07:19 PM Lambda GC = 0.91

2024/05/14/03:07:19 PM Max chi^2 = 25.301

2024/05/14/03:07:19 PM 0 Genome-wide significant SNPs (some may have been

removed by filtering).

2024/05/14/03:07:19 PM

Conversion finished at Tue May 14 15:07:19 2024

2024/05/14/03:07:19 PM Total time elapsed: 28.61s

2024/05/14/03:07:28 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Munging of Trait 2 complete. SNPs remaining:

5099629

2024/05/14/03:07:28 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Warning: The mean chi2 statistic of trait 2 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:07:47 PM Read in Trait 3 summary statistics (5593073 SNPs)

from phen_22_PG_tbn.txt.m.ps ...

2024/05/14/03:07:47 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:47 PM Munging Trait 3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:07:47 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:47 PM Interpreting column names as follows:

2024/05/14/03:07:47 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:07:47 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:07:58 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:08:02 PM Removed 0 SNPs with duplicated rs numbers (5099629

SNPs remain).

2024/05/14/03:08:03 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:08:16 PM Median value of SIGNED_SUMSTAT was 0.00952361,

which seems sensible.

2024/05/14/03:08:17 PM Dropping snps with null values

2024/05/14/03:08:17 PM

Metadata:

2024/05/14/03:08:18 PM Mean chi^2 = 0.951

2024/05/14/03:08:18 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:08:18 PM Lambda GC = 0.934

2024/05/14/03:08:18 PM Max chi^2 = 31.856

2024/05/14/03:08:18 PM 3 Genome-wide significant SNPs (some may have been

removed by filtering).

2024/05/14/03:08:18 PM

Conversion finished at Tue May 14 15:08:18 2024

2024/05/14/03:08:18 PM Total time elapsed: 31.07s

2024/05/14/03:08:29 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:29 PM Munging of Trait 3 complete. SNPs remaining:

5099629

2024/05/14/03:08:29 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:30 PM Warning: The mean chi2 statistic of trait 3 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:08:46 PM Dropped 737230 SNPs due to strand ambiguity,

4362399 SNPs remain in intersection after merging trait1

2024/05/14/03:09:03 PM Dropped 0 SNPs due to strand ambiguity, 4362399

SNPs remain in intersection after merging trait2

2024/05/14/03:09:22 PM Dropped 0 SNPs due to strand ambiguity, 4362399

SNPs remain in intersection after merging trait3

2024/05/14/03:09:22 PM ... Merge of GWAS summary statistics complete.

Number of SNPs: 4362399

2024/05/14/03:09:38 PM Using 4362399 SNPs to estimate Omega (0 SNPs

excluded due to strand ambiguity)

2024/05/14/03:09:38 PM Estimating sigma..

2024/05/14/03:14:29 PM Checking for positive definiteness ..

2024/05/14/03:14:29 PM Sigma hat:

[[1.165 0.97 0.88 ]

[0.97 1.177 0.914]

[0.88 0.914 1.131]]

2024/05/14/03:14:29 PM Mean chi^2 of SNPs used to estimate Omega is low

for some SNPsMTAG may not perform well in this situation.

2024/05/14/03:14:30 PM Beginning estimation of Omega ...

2024/05/14/03:14:30 PM Using GMM estimator of Omega ..

2024/05/14/03:14:31 PM Checking for positive definiteness ..

2024/05/14/03:14:31 PM matrix is not positive definite, performing

adjustment..

2024/05/14/03:14:31 PM Warning: max number of iterations reached in

adjustment procedure. Sigma matrix used is still non-positive-definite.

2024/05/14/03:14:31 PM Completed estimation of Omega ...

2024/05/14/03:14:31 PM Beginning MTAG calculations...

2024/05/14/03:15:19 PM ... Completed MTAG calculations.

2024/05/14/03:15:19 PM Writing Phenotype 1 to file ...

2024/05/14/03:15:58 PM Writing Phenotype 2 to file ...

2024/05/14/03:16:35 PM Writing Phenotype 3 to file ...

2024/05/14/03:17:15 PM

Summary of MTAG results:

Trait # SNPs used N (max) N (mean) GWAS mean chi^2 MTAG mean chi^2 GWAS

equiv. (max) N

1 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.807 18.926 -113599

2 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.792 1.771 -4537

3 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.841 1.444 -3409

Estimated Omega:

[[-1.839e-04 -6.803e-09 -5.303e-09]

[-6.803e-09 -2.001e-04 -5.786e-09]

[-5.303e-09 -5.786e-09 -1.472e-04]]

(Correlation):

[[nan nan nan]

[nan nan nan]

[nan nan nan]]

Estimated Sigma:

[[1.165 0.97 0.88 ]

[0.97 1.177 0.914]

[0.88 0.914 1.131]]

(Correlation):

[[1. 0.828 0.767]

[0.828 1. 0.792]

[0.767 0.792 1. ]]

MTAG weight factors: (average across SNPs)

[-0.014 -0.035 0.061]

2024/05/14/03:17:15 PM

2024/05/14/03:17:15 PM MTAG results saved to file.

2024/05/14/03:17:15 PM MTAG complete. Time elapsed: 11.0m:29.370085001s

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213#issuecomment-2121555944, or

unsubscribe

<

https://github.com/notifications/unsubscribe-auth/AFBUB5KI65WG77QJPAX3243ZDKSJDAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRRGU2TKOJUGQ

.

You are receiving this because you commented.Message ID:

@.***>

—

Reply to this email directly, view it on GitHub, or unsubscribe.

You are receiving this because you authored the thread.Message ID:

@.***>

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213#issuecomment-2124303541, or

unsubscribe

https://github.com/notifications/unsubscribe-auth/AFBUB5OVPXTWXFT7VEGB37LZDRQG3AVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRUGMYDGNJUGE

.

You are receiving this because you commented.Message ID:

@.***>

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: @.***>

paturley commented 3 months ago

1000 may be large for maize samples, but it doesn't appear to be large enough for MTAG, which requires mean chi2 statistics that are greater than one. Sorry I can't be more helpful.

On Wed, May 22, 2024, 7:02 AM real @.***> wrote:

I have more than 1000 samples which are relatively large for maize population.Is there other problems？Thank youTian在 2024年5月22日，18:49，paturley @.***> 写道： Mean chi2 is both a function of heritability and sample size. Your sample

size must be quite small then.

On Wed, May 22, 2024, 5:27 AM real @.***> wrote:

Hello Patrick

Thanks for your reply. I find that the mean chr^2 of my GWAS results for

all the traits were less than 1. My phenotype data includes plant height,

leaf angle and tassel branch numbers, etc which have higher broad sense

heritability.Where could this be the problems?

Thank you and all the best,

Tian

At 2024-05-22 04:40:51, "paturley" @.***> wrote:

Ah. So the problem is that your mean chi2 statistics are less than one, so

the implicit heritability is negative. I generally recommend that MTAG not

be used unless the mean chi2 is at least 1.1 for each trait. Better if it's

even higher than that. In your particular case, I think MTAG may not be the

right tool.

On Mon, May 20, 2024 at 9:53 PM real @.***> wrote:

Hello Patrick

Thanks for your reply. Yes, I think the results for rep_1 is spurious.

The

Manhattan plots for GWAS and MATG as well as the log file were attatched.

Thank you very much!

Best,

Tian

At 2024-05-21 02:41:50, "paturley" @.***> wrote:

Hello Tian,

I'm a bit confused by your question. Are you worried that the very

significant results for rep_1 is spurious? If you post your log file, it

will help me understand what is going on better.

Best,

Patrick

On Mon, May 20, 2024 at 4:06 AM real @.***> wrote:

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of

three

environments of plant hegiht which has low repeatability. We call the

rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG

signials

for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that same

as

the GWAS results. My question are 1) Is MATG suitable for improving the

detection ability of duplicate data？If so, Is rep_1 an abnormal result

and

how to handle it？

Thank you very much!

Tian

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213, or unsubscribe

<

https://github.com/notifications/unsubscribe-auth/AFBUB5LNIGHOATZ7YCQTGRLZDGVHZAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDKMZYGM4DSMI

.

You are receiving this because you are subscribed to this

thread.Message

ID: @.***>

—

Reply to this email directly, view it on GitHub, or unsubscribe.

You are receiving this because you authored the thread.Message ID:

@.***>

2024/05/14/03:05:46 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

<>

<> MTAG: Multi-trait Analysis of GWAS

<> Version: 1.0.8

<> (C) 2017 Omeed Maghzian, Raymond Walters, and Patrick Turley

<> Harvard University Department of Economics / Broad Institute of MIT

and

Harvard

<> GNU General Public License v3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

<> Note: It is recommended to run your own QC on the input before using

this program.

<> Software-related correspondence: @.***

<> All other correspondence: @.***

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Calling ./mtag.py \

--force \

--stream-stdout \

--ld-ref-panel /data/ttian/08-GWAS_1223/01_Emmax_1223/5M_Q_K/LDscores/ \

--sumstats phen_20_PG_tbn.txt.m.ps,phen_21_PG_tbn.txt.m.ps,

phen_22_PG_tbn.txt.m.ps \

--out PG/phen_PG_tbn

phen_20_PG_tbn(rep_1),phen_21_PG_tbn(rep_2),phen_22_PG_tbn(rep_3)

2024/05/14/03:05:46 PM Beginning MTAG analysis...

2024/05/14/03:05:46 PM MTAG will use the Z column for analyses.

2024/05/14/03:05:55 PM Read in Trait 1 summary statistics (5593073 SNPs)

from phen_20_PG_tbn.txt.m.ps ...

2024/05/14/03:05:55 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:05:55 PM Munging Trait 1

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:05:55 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:05:55 PM Interpreting column names as follows:

2024/05/14/03:05:55 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:05:55 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:06:04 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:06:08 PM Removed 0 SNPs with duplicated rs numbers (5099629

SNPs remain).

2024/05/14/03:06:09 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:06:22 PM Median value of SIGNED_SUMSTAT was 0.000275905,

which seems sensible.

2024/05/14/03:06:22 PM Dropping snps with null values

2024/05/14/03:06:22 PM

Metadata:

2024/05/14/03:06:23 PM Mean chi^2 = 0.941

2024/05/14/03:06:23 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:06:23 PM Lambda GC = 0.912

2024/05/14/03:06:23 PM Max chi^2 = 30.569

2024/05/14/03:06:23 PM 1 Genome-wide significant SNPs (some may have been

removed by filtering).

2024/05/14/03:06:23 PM

Conversion finished at Tue May 14 15:06:23 2024

2024/05/14/03:06:23 PM Total time elapsed: 28.06s

2024/05/14/03:06:33 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Munging of Trait 1 complete. SNPs remaining:

5099629

2024/05/14/03:06:33 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Warning: The mean chi2 statistic of trait 1 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:06:51 PM Read in Trait 2 summary statistics (5593073 SNPs)

from phen_21_PG_tbn.txt.m.ps ...

2024/05/14/03:06:51 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:51 PM Munging Trait 2

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:06:51 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:51 PM Interpreting column names as follows:

2024/05/14/03:06:51 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:06:51 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:07:00 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:07:03 PM Removed 0 SNPs with duplicated rs numbers (5099629

SNPs remain).

2024/05/14/03:07:04 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:07:18 PM Median value of SIGNED_SUMSTAT was 0.0122005,

which

seems sensible.

2024/05/14/03:07:18 PM Dropping snps with null values

2024/05/14/03:07:18 PM

Metadata:

2024/05/14/03:07:19 PM Mean chi^2 = 0.933

2024/05/14/03:07:19 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:07:19 PM Lambda GC = 0.91

2024/05/14/03:07:19 PM Max chi^2 = 25.301

2024/05/14/03:07:19 PM 0 Genome-wide significant SNPs (some may have been

removed by filtering).

2024/05/14/03:07:19 PM

Conversion finished at Tue May 14 15:07:19 2024

2024/05/14/03:07:19 PM Total time elapsed: 28.61s

2024/05/14/03:07:28 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Munging of Trait 2 complete. SNPs remaining:

5099629

2024/05/14/03:07:28 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Warning: The mean chi2 statistic of trait 2 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:07:47 PM Read in Trait 3 summary statistics (5593073 SNPs)

from phen_22_PG_tbn.txt.m.ps ...

2024/05/14/03:07:47 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:47 PM Munging Trait 3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:07:47 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:47 PM Interpreting column names as follows:

2024/05/14/03:07:47 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:07:47 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:07:58 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:08:02 PM Removed 0 SNPs with duplicated rs numbers (5099629

SNPs remain).

2024/05/14/03:08:03 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:08:16 PM Median value of SIGNED_SUMSTAT was 0.00952361,

which seems sensible.

2024/05/14/03:08:17 PM Dropping snps with null values

2024/05/14/03:08:17 PM

Metadata:

2024/05/14/03:08:18 PM Mean chi^2 = 0.951

2024/05/14/03:08:18 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:08:18 PM Lambda GC = 0.934

2024/05/14/03:08:18 PM Max chi^2 = 31.856

2024/05/14/03:08:18 PM 3 Genome-wide significant SNPs (some may have been

removed by filtering).

2024/05/14/03:08:18 PM

Conversion finished at Tue May 14 15:08:18 2024

2024/05/14/03:08:18 PM Total time elapsed: 31.07s

2024/05/14/03:08:29 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:29 PM Munging of Trait 3 complete. SNPs remaining:

5099629

2024/05/14/03:08:29 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:30 PM Warning: The mean chi2 statistic of trait 3 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:08:46 PM Dropped 737230 SNPs due to strand ambiguity,

4362399 SNPs remain in intersection after merging trait1

2024/05/14/03:09:03 PM Dropped 0 SNPs due to strand ambiguity, 4362399

SNPs remain in intersection after merging trait2

2024/05/14/03:09:22 PM Dropped 0 SNPs due to strand ambiguity, 4362399

SNPs remain in intersection after merging trait3

2024/05/14/03:09:22 PM ... Merge of GWAS summary statistics complete.

Number of SNPs: 4362399

2024/05/14/03:09:38 PM Using 4362399 SNPs to estimate Omega (0 SNPs

excluded due to strand ambiguity)

2024/05/14/03:09:38 PM Estimating sigma..

2024/05/14/03:14:29 PM Checking for positive definiteness ..

2024/05/14/03:14:29 PM Sigma hat:

[[1.165 0.97 0.88 ]

[0.97 1.177 0.914]

[0.88 0.914 1.131]]

2024/05/14/03:14:29 PM Mean chi^2 of SNPs used to estimate Omega is low

for some SNPsMTAG may not perform well in this situation.

2024/05/14/03:14:30 PM Beginning estimation of Omega ...

2024/05/14/03:14:30 PM Using GMM estimator of Omega ..

2024/05/14/03:14:31 PM Checking for positive definiteness ..

2024/05/14/03:14:31 PM matrix is not positive definite, performing

adjustment..

2024/05/14/03:14:31 PM Warning: max number of iterations reached in

adjustment procedure. Sigma matrix used is still non-positive-definite.

2024/05/14/03:14:31 PM Completed estimation of Omega ...

2024/05/14/03:14:31 PM Beginning MTAG calculations...

2024/05/14/03:15:19 PM ... Completed MTAG calculations.

2024/05/14/03:15:19 PM Writing Phenotype 1 to file ...

2024/05/14/03:15:58 PM Writing Phenotype 2 to file ...

2024/05/14/03:16:35 PM Writing Phenotype 3 to file ...

2024/05/14/03:17:15 PM

Summary of MTAG results:

Trait # SNPs used N (max) N (mean) GWAS mean chi^2 MTAG mean chi^2 GWAS

equiv. (max) N

1 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.807 18.926 -113599

2 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.792 1.771 -4537

3 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.841 1.444 -3409

Estimated Omega:

[[-1.839e-04 -6.803e-09 -5.303e-09]

[-6.803e-09 -2.001e-04 -5.786e-09]

[-5.303e-09 -5.786e-09 -1.472e-04]]

(Correlation):

[[nan nan nan]

[nan nan nan]

[nan nan nan]]

Estimated Sigma:

[[1.165 0.97 0.88 ]

[0.97 1.177 0.914]

[0.88 0.914 1.131]]

(Correlation):

[[1. 0.828 0.767]

[0.828 1. 0.792]

[0.767 0.792 1. ]]

MTAG weight factors: (average across SNPs)

[-0.014 -0.035 0.061]

2024/05/14/03:17:15 PM

2024/05/14/03:17:15 PM MTAG results saved to file.

2024/05/14/03:17:15 PM MTAG complete. Time elapsed: 11.0m:29.370085001s

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213#issuecomment-2121555944, or

unsubscribe

<

https://github.com/notifications/unsubscribe-auth/AFBUB5KI65WG77QJPAX3243ZDKSJDAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRRGU2TKOJUGQ

.

You are receiving this because you commented.Message ID:

@.***>

—

Reply to this email directly, view it on GitHub, or unsubscribe.

You are receiving this because you authored the thread.Message ID:

@.***>

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213#issuecomment-2124303541, or

unsubscribe

< https://github.com/notifications/unsubscribe-auth/AFBUB5OVPXTWXFT7VEGB37LZDRQG3AVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRUGMYDGNJUGE>

.

You are receiving this because you commented.Message ID:

@.***>

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/JonJala/mtag/issues/213#issuecomment-2124514796, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5JO7QKO65BLZN5VBZ3ZDR3KXAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRUGUYTINZZGY . You are receiving this because you commented.Message ID: @.***>

ttian627 commented 3 months ago

OK，Thank you very much.All the best,Tian在 2024年5月22日，19:07，paturley @.***> 写道： 1000 may be large for maize samples, but it doesn't appear to be large

enough for MTAG, which requires mean chi2 statistics that are greater than

one. Sorry I can't be more helpful.

On Wed, May 22, 2024, 7:02 AM real @.***> wrote:

I have more than 1000 samples which are relatively large for maize

population.Is there other problems？Thank youTian在 2024年5月22日，18:49，paturley

@.***> 写道：

Mean chi2 is both a function of heritability and sample size. Your sample

size must be quite small then.

On Wed, May 22, 2024, 5:27 AM real @.***> wrote:

Hello Patrick

Thanks for your reply. I find that the mean chr^2 of my GWAS results for

all the traits were less than 1. My phenotype data includes plant

height,

leaf angle and tassel branch numbers, etc which have higher broad sense

heritability.Where could this be the problems?

Thank you and all the best,

Tian

At 2024-05-22 04:40:51, "paturley" @.***> wrote:

Ah. So the problem is that your mean chi2 statistics are less than one,

so

the implicit heritability is negative. I generally recommend that MTAG

not

be used unless the mean chi2 is at least 1.1 for each trait. Better if

it's

even higher than that. In your particular case, I think MTAG may not be

the

right tool.

On Mon, May 20, 2024 at 9:53 PM real @.***> wrote:

Hello Patrick

Thanks for your reply. Yes, I think the results for rep_1 is spurious.

The

Manhattan plots for GWAS and MATG as well as the log file were

attatched.

Thank you very much!

Best,

Tian

At 2024-05-21 02:41:50, "paturley" @.***> wrote:

Hello Tian,

I'm a bit confused by your question. Are you worried that the very

significant results for rep_1 is spurious? If you post your log file,

it

will help me understand what is going on better.

Best,

Patrick

On Mon, May 20, 2024 at 4:06 AM real @.***> wrote:

Hello,

I used MTAG to enhance the GWAS ability. I have phenotypic data of

three

environments of plant hegiht which has low repeatability. We call

the

rep_1,rep_2,rep_3. The GWAS signials were under 1e-5, but the MTAG

signials

for rep_1 was under 1e-100, rep_2 and rep_3 were under 1e-5 that

same

as

the GWAS results. My question are 1) Is MATG suitable for improving

the

detection ability of duplicate data？If so, Is rep_1 an abnormal

result

and

how to handle it？

Thank you very much!

Tian

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213, or unsubscribe

<

https://github.com/notifications/unsubscribe-auth/AFBUB5LNIGHOATZ7YCQTGRLZDGVHZAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYDKMZYGM4DSMI

.

You are receiving this because you are subscribed to this

thread.Message

ID: @.***>

—

Reply to this email directly, view it on GitHub, or unsubscribe.

You are receiving this because you authored the thread.Message ID:

@.***>

2024/05/14/03:05:46 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

<>

<> MTAG: Multi-trait Analysis of GWAS

<> Version: 1.0.8

<> (C) 2017 Omeed Maghzian, Raymond Walters, and Patrick Turley

<> Harvard University Department of Economics / Broad Institute of MIT

and

Harvard

<> GNU General Public License v3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

<> Note: It is recommended to run your own QC on the input before

using

this program.

<> Software-related correspondence: @.***

<> All other correspondence: @.***

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

Calling ./mtag.py \

--force \

--stream-stdout \

--ld-ref-panel /data/ttian/08-GWAS_1223/01_Emmax_1223/5M_Q_K/LDscores/

\

--sumstats phen_20_PG_tbn.txt.m.ps,phen_21_PG_tbn.txt.m.ps,

phen_22_PG_tbn.txt.m.ps \

--out PG/phen_PG_tbn

phen_20_PG_tbn(rep_1),phen_21_PG_tbn(rep_2),phen_22_PG_tbn(rep_3)

2024/05/14/03:05:46 PM Beginning MTAG analysis...

2024/05/14/03:05:46 PM MTAG will use the Z column for analyses.

2024/05/14/03:05:55 PM Read in Trait 1 summary statistics (5593073

SNPs)

from phen_20_PG_tbn.txt.m.ps ...

2024/05/14/03:05:55 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:05:55 PM Munging Trait 1

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:05:55 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:05:55 PM Interpreting column names as follows:

2024/05/14/03:05:55 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:05:55 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:06:04 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous

SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:06:08 PM Removed 0 SNPs with duplicated rs numbers

(5099629

SNPs remain).

2024/05/14/03:06:09 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:06:22 PM Median value of SIGNED_SUMSTAT was 0.000275905,

which seems sensible.

2024/05/14/03:06:22 PM Dropping snps with null values

2024/05/14/03:06:22 PM

Metadata:

2024/05/14/03:06:23 PM Mean chi^2 = 0.941

2024/05/14/03:06:23 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:06:23 PM Lambda GC = 0.912

2024/05/14/03:06:23 PM Max chi^2 = 30.569

2024/05/14/03:06:23 PM 1 Genome-wide significant SNPs (some may have

been

removed by filtering).

2024/05/14/03:06:23 PM

Conversion finished at Tue May 14 15:06:23 2024

2024/05/14/03:06:23 PM Total time elapsed: 28.06s

2024/05/14/03:06:33 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Munging of Trait 1 complete. SNPs remaining:

5099629

2024/05/14/03:06:33 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:33 PM Warning: The mean chi2 statistic of trait 1 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:06:51 PM Read in Trait 2 summary statistics (5593073

SNPs)

from phen_21_PG_tbn.txt.m.ps ...

2024/05/14/03:06:51 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:51 PM Munging Trait 2

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:06:51 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:06:51 PM Interpreting column names as follows:

2024/05/14/03:06:51 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:06:51 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:07:00 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous

SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:07:03 PM Removed 0 SNPs with duplicated rs numbers

(5099629

SNPs remain).

2024/05/14/03:07:04 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:07:18 PM Median value of SIGNED_SUMSTAT was 0.0122005,

which

seems sensible.

2024/05/14/03:07:18 PM Dropping snps with null values

2024/05/14/03:07:18 PM

Metadata:

2024/05/14/03:07:19 PM Mean chi^2 = 0.933

2024/05/14/03:07:19 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:07:19 PM Lambda GC = 0.91

2024/05/14/03:07:19 PM Max chi^2 = 25.301

2024/05/14/03:07:19 PM 0 Genome-wide significant SNPs (some may have

been

removed by filtering).

2024/05/14/03:07:19 PM

Conversion finished at Tue May 14 15:07:19 2024

2024/05/14/03:07:19 PM Total time elapsed: 28.61s

2024/05/14/03:07:28 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Munging of Trait 2 complete. SNPs remaining:

5099629

2024/05/14/03:07:28 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:28 PM Warning: The mean chi2 statistic of trait 2 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:07:47 PM Read in Trait 3 summary statistics (5593073

SNPs)

from phen_22_PG_tbn.txt.m.ps ...

2024/05/14/03:07:47 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:47 PM Munging Trait 3

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><

2024/05/14/03:07:47 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:07:47 PM Interpreting column names as follows:

2024/05/14/03:07:47 PM snpid: Variant ID (e.g., rs number)

n: Sample size

a1: a1, interpreted as ref allele for signed sumstat.

pval: p-Value

a2: a2, interpreted as non-ref allele for signed sumstat.

z: Directional summary statistic as specified by --signed-sumstats.

2024/05/14/03:07:47 PM Reading sumstats from provided DataFrame into

memory 10000000 SNPs at a time.

2024/05/14/03:07:58 PM Read 5593073 SNPs from --sumstats file.

Removed 0 SNPs with missing values.

Removed 0 SNPs with INFO <= None.

Removed 0 SNPs with MAF <= 0.01.

Removed 0 SNPs with SE <0 or NaN values.

Removed 0 SNPs with out-of-bounds p-values.

Removed 493444 variants that were not SNPs. Note: strand ambiguous

SNPs

were not dropped.

5099629 SNPs remain.

2024/05/14/03:08:02 PM Removed 0 SNPs with duplicated rs numbers

(5099629

SNPs remain).

2024/05/14/03:08:03 PM Removed 0 SNPs with N < 815.333333333 (5099629

SNPs

remain).

2024/05/14/03:08:16 PM Median value of SIGNED_SUMSTAT was 0.00952361,

which seems sensible.

2024/05/14/03:08:17 PM Dropping snps with null values

2024/05/14/03:08:17 PM

Metadata:

2024/05/14/03:08:18 PM Mean chi^2 = 0.951

2024/05/14/03:08:18 PM WARNING: mean chi^2 may be too small.

2024/05/14/03:08:18 PM Lambda GC = 0.934

2024/05/14/03:08:18 PM Max chi^2 = 31.856

2024/05/14/03:08:18 PM 3 Genome-wide significant SNPs (some may have

been

removed by filtering).

2024/05/14/03:08:18 PM

Conversion finished at Tue May 14 15:08:18 2024

2024/05/14/03:08:18 PM Total time elapsed: 31.07s

2024/05/14/03:08:29 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:29 PM Munging of Trait 3 complete. SNPs remaining:

5099629

2024/05/14/03:08:29 PM

<><><<>><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

2024/05/14/03:08:30 PM Warning: The mean chi2 statistic of trait 3 is

less

1.02 - MTAG estimates may be unstable.

2024/05/14/03:08:46 PM Dropped 737230 SNPs due to strand ambiguity,

4362399 SNPs remain in intersection after merging trait1

2024/05/14/03:09:03 PM Dropped 0 SNPs due to strand ambiguity, 4362399

SNPs remain in intersection after merging trait2

2024/05/14/03:09:22 PM Dropped 0 SNPs due to strand ambiguity, 4362399

SNPs remain in intersection after merging trait3

2024/05/14/03:09:22 PM ... Merge of GWAS summary statistics complete.

Number of SNPs: 4362399

2024/05/14/03:09:38 PM Using 4362399 SNPs to estimate Omega (0 SNPs

excluded due to strand ambiguity)

2024/05/14/03:09:38 PM Estimating sigma..

2024/05/14/03:14:29 PM Checking for positive definiteness ..

2024/05/14/03:14:29 PM Sigma hat:

[[1.165 0.97 0.88 ]

[0.97 1.177 0.914]

[0.88 0.914 1.131]]

2024/05/14/03:14:29 PM Mean chi^2 of SNPs used to estimate Omega is

low

for some SNPsMTAG may not perform well in this situation.

2024/05/14/03:14:30 PM Beginning estimation of Omega ...

2024/05/14/03:14:30 PM Using GMM estimator of Omega ..

2024/05/14/03:14:31 PM Checking for positive definiteness ..

2024/05/14/03:14:31 PM matrix is not positive definite, performing

adjustment..

2024/05/14/03:14:31 PM Warning: max number of iterations reached in

adjustment procedure. Sigma matrix used is still

non-positive-definite.

2024/05/14/03:14:31 PM Completed estimation of Omega ...

2024/05/14/03:14:31 PM Beginning MTAG calculations...

2024/05/14/03:15:19 PM ... Completed MTAG calculations.

2024/05/14/03:15:19 PM Writing Phenotype 1 to file ...

2024/05/14/03:15:58 PM Writing Phenotype 2 to file ...

2024/05/14/03:16:35 PM Writing Phenotype 3 to file ...

2024/05/14/03:17:15 PM

Summary of MTAG results:

Trait # SNPs used N (max) N (mean) GWAS mean chi^2 MTAG mean chi^2

GWAS

equiv. (max) N

1 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.807 18.926 -113599

2 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.792 1.771 -4537

3 ..._PG_tbn.txt.m.ps 4362399 1223 1223 0.841 1.444 -3409

Estimated Omega:

[[-1.839e-04 -6.803e-09 -5.303e-09]

[-6.803e-09 -2.001e-04 -5.786e-09]

[-5.303e-09 -5.786e-09 -1.472e-04]]

(Correlation):

[[nan nan nan]

[nan nan nan]

[nan nan nan]]

Estimated Sigma:

[[1.165 0.97 0.88 ]

[0.97 1.177 0.914]

[0.88 0.914 1.131]]

(Correlation):

[[1. 0.828 0.767]

[0.828 1. 0.792]

[0.767 0.792 1. ]]

MTAG weight factors: (average across SNPs)

[-0.014 -0.035 0.061]

2024/05/14/03:17:15 PM

2024/05/14/03:17:15 PM MTAG results saved to file.

2024/05/14/03:17:15 PM MTAG complete. Time elapsed:

11.0m:29.370085001s

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213#issuecomment-2121555944,

or

unsubscribe

<

https://github.com/notifications/unsubscribe-auth/AFBUB5KI65WG77QJPAX3243ZDKSJDAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRRGU2TKOJUGQ

.

You are receiving this because you commented.Message ID:

@.***>

—

Reply to this email directly, view it on GitHub, or unsubscribe.

You are receiving this because you authored the thread.Message ID:

@.***>

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213#issuecomment-2124303541,

or

unsubscribe

<

https://github.com/notifications/unsubscribe-auth/AFBUB5OVPXTWXFT7VEGB37LZDRQG3AVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRUGMYDGNJUGE>

.

You are receiving this because you commented.Message ID:

@.***>

—Reply to this email directly, view it on GitHub, or unsubscribe.You are

receiving this because you authored the thread.Message ID: @.***>

—

Reply to this email directly, view it on GitHub

https://github.com/JonJala/mtag/issues/213#issuecomment-2124514796, or

unsubscribe

https://github.com/notifications/unsubscribe-auth/AFBUB5JO7QKO65BLZN5VBZ3ZDR3KXAVCNFSM6AAAAABH7H4OY2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMRUGUYTINZZGY

.

You are receiving this because you commented.Message ID:

@.***>

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: @.***>