Blank meta result for centain phenotypes

[JASMINE0926]

Hi everyone,

I am working with two datasets, one European and one Asian, and have successfully calculated LD scores. I have a total of 21 phenotypes. For most phenotypes, the meta-analysis ran successfully and yielded significant loci. However, I encountered an issue with some phenotypes where the meta-analysis results were empty. For instance, phenotype ID 228 resulted in an empty meta-analysis, while phenotype ID 227 calculated normally. I have attached the log files for both phenotypes for reference.

I would appreciate any insights into what might be causing this issue. Thank you in advance for your help!

Best regards,
Jasmine
227_MAMA.log
228_MAMA.log

[paturley]

Hi Jasmine, It looks like from your logs that you have a negative coefficient associated with the LD score for the EAS population for that phenotype. This implies a negative estimate of heritability, which causes the MAMA method to break down. While I can't tell this from looking at your log, this is more likely to happen when you are looking at a GWAS that has low power. Do you know what the mean chi2 stat is for that ancestry/phenotype? Patrick

…

On Mon, Feb 19, 2024 at 3:25 AM jasmine ***@***.***> wrote: Hi everyone, I am working with two datasets, one European and one Asian, and have successfully calculated LD scores. I have a total of 21 phenotypes. For most phenotypes, the meta-analysis ran successfully and yielded significant loci. However, I encountered an issue with some phenotypes where the meta-analysis results were empty. For instance, phenotype ID 227 resulted in an empty meta-analysis, while phenotype ID 228 calculated normally. I have attached the log files for both phenotypes for reference. I would appreciate any insights into what might be causing this issue. Thank you in advance for your help! Best regards, Jasmine 227_MAMA.log <https://github.com/JonJala/mama/files/14328449/227_MAMA.log> 228_MAMA.log <https://github.com/JonJala/mama/files/14328450/228_MAMA.log> — Reply to this email directly, view it on GitHub <#35>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFBUB5O7O3JKV3WZ4GR6VM3YUMEAFAVCNFSM6AAAAABDPAI5V6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2DCNZZGIYTQNI> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[JASMINE0926]

Dear Patrick,

Thank you for your prompt response and insightful comments. In total, I have 21 phenotypes, among which phenotypes with IDs 228, 229, 237, and 239 failed to calculate in MAMA. The data for these phenotypes in the EAS population are as follows:

Phenotype Total Observed scale h2 Lambda GC Mean Chi2 Intercept Ratio
228 0.0929 (0.0645) 1.0195 1.0188 1.0051 0.2718
229 0.1513 (0.0587) 1.0225 1.022 0.9999 < 0
237 -0.0107 (0.0591) 1.0046 1.001 1.0025 2.5645
239 0.0411 (0.0588) 1.0046 1.0134 1.0075 0.5612
These phenotypes seem to have issues, especially with a negative estimate of heritability for phenotype 237. I would appreciate any further insights or suggestions you might have on how to address these problems.

Best regards,

Jasmine

Hi Jasmine, It looks like from your logs that you have a negative coefficient associated with the LD score for the EAS population for that phenotype. This implies a negative estimate of heritability, which causes the MAMA method to break down. While I can't tell this from looking at your log, this is more likely to happen when you are looking at a GWAS that has low power. Do you know what the mean chi2 stat is for that ancestry/phenotype? Patrick
…
On Mon, Feb 19, 2024 at 3:25 AM jasmine @.> wrote: Hi everyone, I am working with two datasets, one European and one Asian, and have successfully calculated LD scores. I have a total of 21 phenotypes. For most phenotypes, the meta-analysis ran successfully and yielded significant loci. However, I encountered an issue with some phenotypes where the meta-analysis results were empty. For instance, phenotype ID 227 resulted in an empty meta-analysis, while phenotype ID 228 calculated normally. I have attached the log files for both phenotypes for reference. I would appreciate any insights into what might be causing this issue. Thank you in advance for your help! Best regards, Jasmine 227_MAMA.log https://github.com/JonJala/mama/files/14328449/227_MAMA.log 228_MAMA.log https://github.com/JonJala/mama/files/14328450/228_MAMA.log — Reply to this email directly, view it on GitHub <#35>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5O7O3JKV3WZ4GR6VM3YUMEAFAVCNFSM6AAAAABDPAI5V6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2DCNZZGIYTQNI . You are receiving this because you are subscribed to this thread.Message ID: @.>

Hi Jasmine, It looks like from your logs that you have a negative coefficient associated with the LD score for the EAS population for that phenotype. This implies a negative estimate of heritability, which causes the MAMA method to break down. While I can't tell this from looking at your log, this is more likely to happen when you are looking at a GWAS that has low power. Do you know what the mean chi2 stat is for that ancestry/phenotype? Patrick
…
On Mon, Feb 19, 2024 at 3:25 AM jasmine @.> wrote: Hi everyone, I am working with two datasets, one European and one Asian, and have successfully calculated LD scores. I have a total of 21 phenotypes. For most phenotypes, the meta-analysis ran successfully and yielded significant loci. However, I encountered an issue with some phenotypes where the meta-analysis results were empty. For instance, phenotype ID 227 resulted in an empty meta-analysis, while phenotype ID 228 calculated normally. I have attached the log files for both phenotypes for reference. I would appreciate any insights into what might be causing this issue. Thank you in advance for your help! Best regards, Jasmine 227_MAMA.log https://github.com/JonJala/mama/files/14328449/227_MAMA.log 228_MAMA.log https://github.com/JonJala/mama/files/14328450/228_MAMA.log — Reply to this email directly, view it on GitHub <#35>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5O7O3JKV3WZ4GR6VM3YUMEAFAVCNFSM6AAAAABDPAI5V6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2DCNZZGIYTQNI . You are receiving this because you are subscribed to this thread.Message ID: @.>

[paturley]

Hi Jasmine, I'm not sure what a reasonable solution is in this case. MAMA generally requires some baseline levels of power to estimate some key parameters (e.g., the heritability), and if the GWASs are too small to get reasonable estimates of those parameters, then MAMA is pretty unreliable. The only way to get MAMA to run is to force MAMA to make assumptions about the heritability and genetic correlation between traits, but you may not want to do that if you don't have evidence. Sorry I can't be more helpful. Best, Patrick

…

On Thu, Feb 22, 2024 at 7:17 AM jasmine ***@***.***> wrote: Dear Patrick, Thank you for your prompt response and insightful comments. In total, I have 21 phenotypes, among which phenotypes with IDs 228, 229, 237, and 239 failed to calculate in MAMA. The data for these phenotypes in the EAS population are as follows: Phenotype Total Observed scale h2 Lambda GC Mean Chi2 Intercept Ratio 228 0.0929 (0.0645) 1.0195 1.0188 1.0051 0.2718 229 0.1513 (0.0587) 1.0225 1.022 0.9999 < 0 237 -0.0107 (0.0591) 1.0046 1.001 1.0025 2.5645 239 0.0411 (0.0588) 1.0046 1.0134 1.0075 0.5612 These phenotypes seem to have issues, especially with a negative estimate of heritability for phenotype 237. I would appreciate any further insights or suggestions you might have on how to address these problems. Best regards, Jasmine Hi Jasmine, It looks like from your logs that you have a negative coefficient associated with the LD score for the EAS population for that phenotype. This implies a negative estimate of heritability, which causes the MAMA method to break down. While I can't tell this from looking at your log, this is more likely to happen when you are looking at a GWAS that has low power. Do you know what the mean chi2 stat is for that ancestry/phenotype? Patrick … <#m_493866027039483776_> On Mon, Feb 19, 2024 at 3:25 AM jasmine *@*.*> wrote: Hi everyone, I am working with two datasets, one European and one Asian, and have successfully calculated LD scores. I have a total of 21 phenotypes. For most phenotypes, the meta-analysis ran successfully and yielded significant loci. However, I encountered an issue with some phenotypes where the meta-analysis results were empty. For instance, phenotype ID 227 resulted in an empty meta-analysis, while phenotype ID 228 calculated normally. I have attached the log files for both phenotypes for reference. I would appreciate any insights into what might be causing this issue. Thank you in advance for your help! Best regards, Jasmine 227_MAMA.log https://github.com/JonJala/mama/files/14328449/227_MAMA.log <https://github.com/JonJala/mama/files/14328449/227_MAMA.log> 228_MAMA.log https://github.com/JonJala/mama/files/14328450/228_MAMA.log <https://github.com/JonJala/mama/files/14328450/228_MAMA.log> — Reply to this email directly, view it on GitHub <#35 <#35>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5O7O3JKV3WZ4GR6VM3YUMEAFAVCNFSM6AAAAABDPAI5V6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2DCNZZGIYTQNI <https://github.com/notifications/unsubscribe-auth/AFBUB5O7O3JKV3WZ4GR6VM3YUMEAFAVCNFSM6AAAAABDPAI5V6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2DCNZZGIYTQNI> . You are receiving this because you are subscribed to this thread.Message ID: @.*> Hi Jasmine, It looks like from your logs that you have a negative coefficient associated with the LD score for the EAS population for that phenotype. This implies a negative estimate of heritability, which causes the MAMA method to break down. While I can't tell this from looking at your log, this is more likely to happen when you are looking at a GWAS that has low power. Do you know what the mean chi2 stat is for that ancestry/phenotype? Patrick … <#m_493866027039483776_> On Mon, Feb 19, 2024 at 3:25 AM jasmine *@*.*> wrote: Hi everyone, I am working with two datasets, one European and one Asian, and have successfully calculated LD scores. I have a total of 21 phenotypes. For most phenotypes, the meta-analysis ran successfully and yielded significant loci. However, I encountered an issue with some phenotypes where the meta-analysis results were empty. For instance, phenotype ID 227 resulted in an empty meta-analysis, while phenotype ID 228 calculated normally. I have attached the log files for both phenotypes for reference. I would appreciate any insights into what might be causing this issue. Thank you in advance for your help! Best regards, Jasmine 227_MAMA.log https://github.com/JonJala/mama/files/14328449/227_MAMA.log <https://github.com/JonJala/mama/files/14328449/227_MAMA.log> 228_MAMA.log https://github.com/JonJala/mama/files/14328450/228_MAMA.log <https://github.com/JonJala/mama/files/14328450/228_MAMA.log> — Reply to this email directly, view it on GitHub <#35 <#35>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFBUB5O7O3JKV3WZ4GR6VM3YUMEAFAVCNFSM6AAAAABDPAI5V6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2DCNZZGIYTQNI <https://github.com/notifications/unsubscribe-auth/AFBUB5O7O3JKV3WZ4GR6VM3YUMEAFAVCNFSM6AAAAABDPAI5V6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGE2DCNZZGIYTQNI> . You are receiving this because you are subscribed to this thread.Message ID: @.*> — Reply to this email directly, view it on GitHub <#35 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFBUB5KXWJRWTO3SV6IFDHLYU4ZOBAVCNFSM6AAAAABDPAI5V6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNJZGMZTSMBZGY> . You are receiving this because you commented.Message ID: ***@***.***>