FTND meta-analysis wave3

[jaamarks]

Update Nicotine dependence meta-analysis.

[jaamarks]

Hi all, meta-analyses are complete—details below.

Cross Ancestry

This meta-analysis included 31 groups:

AAND_AA, ADAA_AA, COGEND2_AA, COGEND2_EA, COGEND_AA, COGEND_EA, COPDGENE1_AA,
COPDGENE1_EA, COPDGENE2_AA, COPDGENE2_EA, DECODE_EA, DENTAL_CARIES_EA,
EAGLE_EA, EMERGE_EA, FINRISK_EA, FTC_EA, GAIN_AA, GAIN_EA, GERMAN_EA,
JHS_ARIC_AA, MINNESOTA_TWINS_EA, NELSON_EA, NONGAIN_EA, NTR_EA, S4S_EA,
SAGE_AA, SAGE_EA, UW_TTURC_AA, UW_TTURC_EA, YALE_PENN_AA, YALE_PENN_EA

Mr. Big meta results:

MarkerName,chr,position,Allele1,Allele2,Effect,StdErr,P.value,Direction,HetISq,HetChiSq,HetDf,HetPVal
rs16969968:78882925:G:A,15,78882925,a,g,0.0604,0.0046,1.795e-39,++++++++-++++++++++-+++++++-+++,38.9,49.102,30,0.01535

Plots with applied filters: MAF<0.01, RSQ<0.30, remove singletons (SNPs only present in one cohort)

---

EA-specific

This meta-analysis included 20 groups:

COGEND2_EA, COGEND_EA, COPDGENE1_EA, COPDGENE2_EA, DECODE_EA, DENTAL_CARIES_EA, 
EAGLE_EA, EMERGE_EA, FINRISK_EA, FTC_EA, GAIN_EA, GERMAN_EA, MINNESOTA_TWINS_EA, 
NELSON_EA, NONGAIN_EA, NTR_EA, S4S_EA, SAGE_EA, UW_TTURC_EA, YALE_PENN_EA

Mr. Big meta results:

MarkerName,chr,position,Allele1,Allele2,Effect,StdErr,P.value,Direction,HetISq,HetChiSq,HetDf,HetPVal
rs16969968:78882925:G:A,15,78882925,a,g,0.0611,0.0047,5.648e-38,++++++++++++++++++++,36.9,30.126,19,0.05022

Plots with applied filters: MAF<0.01, RSQ<0.30, remove singletons (SNPs only present in one cohort)

---

AA-specific

This meta-analysis included 11 groups: NOTE, ADAA_AA was updated in this meta-analysis to included the correct beta orientation.

AAND_AA, ADAA_AA, COGEND2_AA, COGEND_AA, COPDGENE1_AA, COPDGENE2_AA, GAIN_AA, 
JHS_ARIC_AA, SAGE_AA, UW_TTURC_AA, YALE_PENN_AA

Mr. Big meta results:

MarkerName,chr,position,Allele1,Allele2,Effect,StdErr,P.value,Direction,HetISq,HetChiSq,HetDf,HetPVal
rs16969968:78882925:G:A 15 78882925 a g 0.0493 0.0183 0.00709 +++++-+-+-+ 46.2 18.584 10 0.04588 

Plots with applied filters: MAF<0.01, RSQ<0.30, remove singletons (SNPs only present in one cohort)

[jaamarks]

Cross Ancestry top 5 SNPs

chr20

MarkerName chr P.value
rs151176846:61997500:T:C 20 1.304e-12
rs6011779:61984317:C:T 20 7.802e-12
rs45497800:61991833:C:T 20 1.384e-11
rs12479680:61988413:A:G 20 2.48e-11
rs6062899:61979793:G:A 20 3.986e-11

chr15

MarkerName chr P.value
rs16969968:78882925:G:A 15 1.795e-39
rs146009840:78906177:A:T 15 2.969e-39
rs8192482:78886198:C:T 15 3.568e-39
rs8034191:78806023:T:C 15 4.81e-39
rs4887067:78886947:G:A 15 9.067e-39

chr9

MarkerName chr P.value
rs13284520:136502572:A:C 9 3.513e-08
rs3025385:136502739:C:T 9 5.002e-08
rs3025387:136502987:A:G 9 5.414e-08
rs3025386:136502764:T:C 9 5.522e-08
rs3025388:136503256:A:G 9 7.748e-08

chr7

MarkerName chr P.value
rs2714700:79367667:T:C 7 2.543e-08
rs2714674:79385250:C:T 7 3.002e-08
rs1464692:79378243:G:A 7 1.084e-07
rs2707864:79403735:T:C 7 2.816e-07
rs60613294:57849672:G:GTGTTTGTT 7 1.073e-06

chr5

MarkerName chr P.value
rs1862416:167394595:T:C 5 3.473e-08
rs36064369:167396567:G:C 5 6.463e-08
rs116612101:167383503:C:T 5 2.261e-07
rs16892055:33977004:C:T 5 3.827e-07
rs1862415:167394921:C:T 5 6.15e-07

20181030_ftnd_meta_analysis_wave3.afr+eur.exclude_singletons.1df.p_lte_0.001.csv.gz

[jaamarks]

For updates, see https://github.com/RTIInternational/bioinformatics/issues/55#issuecomment-437125664

This update includes the correct ADAA_AA beta orientation. We also fixed the formatting on the German_EA cohort that was causing the merging issues. All of the results have been uploaded to S3 at: s3://rti-nd/META/1df/20181108

[jaamarks]

dbGaP Submission

April 27, 2020: Received an email from Neha Gupta (NCBI):

I am processing analysis data for study phs001532.v2.p1. Your meta-info sheet is missing some required information. Please check the Association data dbGaP template attached in the email and please provide all the relevant information for your data and specially the required fields.

Here was the file they attached:
Association_Analysis.xlsx

---

April 30, 2020:

• I forwarded this to Dana and promised to look into it.
• I asked for clarification from Neha about the missing data.
• Neha emails back with some clarifications.
  • The submission was missing some required fields in the meta-analysis results. In particular missing was the Effect allele column (which is actually our Allele1) and the Effect allele frequency.
  • Neha also sent a screen shot of the meta-info sheet received from the dbGaP submission:

• Here is what Neha showed was missing (outlined in blue):

---

May 8, 2020: EUR-AFR meta-analysis reran with the added field: Effect allele frequency (Freq1) and uploaded to S3
May 12, 2020: AFR meta-analyses reran and uploaded to S3
May 15, 2020:

• Neha sends follow-up email.
• I reply that I am working on finishing up the analyses

May 18, 2020: the EUR were upload to S3
May 20, 2020: I email Neha back saying I complete the metas. I also attached one of the meta-info sheets and let them know that it appeared on my end that we included all of the required info in this sheet including the sections they had outlined in blue. I say that I will upload the new results if all looks good.

May 21, 2020:

• Dana asks if I specified a release date. I reply that I did not.
• Went ahead and uploaded the new meta-analyses and the info sheets with updated titles
  • Need to clarify that the titles are good-to-go or if they should be the same as the title of the paper submitted to biorXiv.
• after reading more closely in the meta-analysis tab of the template spreadsheet, it appears that we need to reformat our GWAS meta-analysis results files. In particular they need to have the column headers labeled to match their template outlined in the meta-analysis tab. This means that we will have a bit of reformatting to do other than just renaming the headers.
  • Change SNP_ID from (rsid:pos:a1:a2) to rsid.
  • what to do about SNPs without rsID? keep as chr:pos:a1:a2
  • capture the a1 and a2 from the SNP_ID and create new columns with these (ref and alt given by dbSNP)
  • rename current Allele1 to Effect allele
  • remove current columns labeled: Allele2, FreqSE, Direction, HetISq, HetChiSq, HetDf, HetPVal