Unmodifiable Conditional Analysis pipeline

Summary

Unmodifiable pipelines are predefined workflows that cannot be modified by the user. The advantage of running unmodifiable pipelines compared to modifiable pipelines is that you will get results directly to the green library and the User results PheWeb browser. No download requests are needed, because results of unmodifiable pipeline have been verified not to contain any individual-level data. Running the finemapping unmodifiable pipeline is very similar to running finemapping in the modifiable pipelines, with some small restrictions. The unmodifiable finemapping pipeline can be accessed in the sandbox from The pipelines app -> unmodifable workflow -> Unmodifiable Conditional DF12/13F13). For more information about the Pipelines tool, check the Pipelines tool documentation.

How it Works

The pipeline first take the input sumstats and runs the finemap pipeline to extract regions of interest. Then a step filters out regions where hits > locus_mlogp_threshold .

Then, regenie is called for each of these regions and rounds of conditional analysis are run until either the top hit is below the conditioning_mlogp_threshold or max_steps of iterations are reached.

Summary of inputs

• regenie_cond_sb.sumstats_root: path to the sumstats. PHENO will be replaced by the pheno parameter.

• regenie_cond_sb.prefix: the prefix of the output

• regenie_cond_sb.pheno: Phenotype name. It needs to match the sumstats_root basename

• regenie_cond_sb.regenie_conditional.null: Path to Loco of regenie step1.

• regenie_cond_sb.pheno_file: Path to pheno file.

• regenie_cond_sb.cov_file: Path to cov file.

• regenie_cond_sb.is_binary: "true" or "false". it tells regenie whether the pheno is binary or not

• regenie_cond_sb.covariates: List of covariates. The pipeline will filter them out to make sure only covaraites with >5 samples are included

• regenie_cond_sb.chroms: Comma separated list of chroms to include.

• regenie_cond_sb.regenie_conditional.max_steps: Maximum number of iterations in the conditional analysis

• regenie_cond_sb.locus_mlogp_threshold: Mlogp threshold for the initial hit

• regenie_cond_sb.conditioning_mlogp_threshold: Mlogp threshold for the conditioned hit

• These are options for the finemapping threshold. They can be changed for users are familiar with the pipeline, but we recommend not touching them

  • regenie_cond_sb.finemap_regions.rsid_col: ""

  • regenie_cond_sb.finemap_regions.x_chromosome: "true"

  • regenie_cond_sb.finemap_regions.window: "1500000"

  • regenie_cond_sb.finemap_regions.scale_se_by_pval: "false"

  • regenie_cond_sb.finemap_regions.p_threshold: "0.00000005"

  • regenie_cond_sb.finemap_regions.max_region_width: "10000000"

  • regenie_cond_sb.finemap_regions.window_shrink_ratio: "0.9"

Outputs

There are two output files.

• all_hits . A file that merges all pheno chains. This file contains all the steps done for each chain. In this case the first hits did not cause a chain as the new top hits in turn did not have significat pvalues. E.g.

  VARIANT	BETA	SE	MLOG10P	BETA_cond	SE_cond	MLOG10P_cond	VARIANT_cond
  chr1_17381854_G_A	-0.0371	0.0064	8.2425	nan	nan	nan	nan
  chr2_161145756_T_A	-0.0429	0.0073	8.3533	nan	nan	nan	nan
  chr2_196338002_G_A	0.059	0.0095	9.2624	nan	nan	nan	nan
  chr3_18633152_A_C	0.044	0.0067	10.3279	nan	nan	nan	nan
  chr3_16811234_G_A	-0.0367	0.0065	7.7032	-0.0328	0.0066	6.1828	chr3_18633152_A_C
  chr3_43656252_G_A	0.0759	0.01	13.5747	nan	nan	nan	nan
  chr3_44176307_AAAG_A	0.7812	0.1679	5.4822	0.9816	0.1974	6.1793	chr3_43656252_G_A
  chr3_158237129_A_G	0.0388	0.0065	8.6487	nan	nan	nan	nan
  chr3_177171295_G_A	0.0843	0.0133	9.588	nan	nan	nan	nan
  chr4_59388159_A_G	-0.0403	0.0071	7.8402	nan	nan	nan	nan
  chr5_104451467_A_C	0.0383	0.0066	8.1869	nan	nan	nan	nan
  chr5_120730291_A_G	-0.0437	0.0066	10.5128	nan	nan	nan	nan
  chr5_121149217_A_C	0.6199	0.1257	6.086	0.7808	0.1422	7.3998	chr5_120730291_A_G
  chr5_144466903_T_C	0.0374	0.0065	8.1641	nan	nan	nan	nan
  chr5_153777668_A_C	-0.0381	0.0065	8.3132	nan	nan	nan	nan
  chr5_153129992_T_G	-0.0329	0.0065	6.3568	-0.032	0.0065	6.0592	chr5_153777668_A_C
  chr6_51715850_T_C	-0.0567	0.0103	7.4145	nan	nan	nan	nan
  chr6_151880872_T_C	-0.0474	0.0065	12.4306	nan	nan	nan	nan
  chr6_164751626_G_A	-0.038	0.0063	8.717	nan	nan	nan	nan
  chr9_36999372_C_T	0.0348	0.0063	7.419	nan	nan	nan	nan
  chr9_95537395_T_A	0.0565	0.0102	7.4833	nan	nan	nan	nan
  chr10_21600368_A_AAC	-0.0515	0.0085	8.819	nan	nan	nan	nan
  chr10_104842738_C_CT	-0.0411	0.007	8.2919	nan	nan	nan	nan
  chr11_28582122_A_G	-0.0389	0.0066	8.3234	nan	nan	nan	nan
  chr11_99647795_T_C	0.0431	0.0069	9.5016	nan	nan	nan	nan
  chr11_113324873_A_G	-0.038	0.0064	8.5837	nan	nan	nan	nan
  chr11_115118275_G_A	-0.0407	0.007	8.312	-0.0398	0.007	7.9104	chr11_113324873_A_G
  chr11_113050349_C_T	0.0364	0.0064	7.8798	0.0331	0.0064	6.5628	chr11_113324873_A_G,chr11_115118275_G_A
  chr12_60397384_T_A	-0.0372	0.0065	8.0402	nan	nan	nan	nan
  chr14_70749996_T_C	-0.227	0.0359	9.5829	nan	nan	nan	nan
  chr15_73812581_C_A	-0.0584	0.0105	7.6144	nan	nan	nan	nan
  chr16_21667465_T_C	0.0434	0.0073	8.6185	nan	nan	nan	nan
  chr16_65751219_A_G	0.0357	0.0064	7.614	nan	nan	nan	nan
  chr16_72134342_T_G	0.0485	0.0063	13.8185	nan	nan	nan	nan
  chr16_71324662_TCTGCTAATAGGAGTGA_T	0.0829	0.0128	10.054	0.0728	0.0131	7.582	chr16_72134342_T_G
  chr18_53199573_T_C	0.0497	0.0063	14.3777	nan	nan	nan	nan
  chr18_55212207_T_C	0.0778	0.0148	6.843	0.0759	0.015	6.3537	chr18_53199573_T_C
  chr18_76052843_T_C	0.0474	0.0085	7.6577	nan	nan	nan	nan
  chr19_49457409_G_T	0.0795	0.0136	8.3038	nan	nan	nan	nan
  chrX_93697357_A_G	0.4882	0.0889	7.3997	nan	nan	nan	nan
  chrX_93934138_A_G	0.7061	0.1479	5.7454	1.0262	0.1883	7.2998	chrX_93697357_A_G

  chr3_18633152_A_C is the first hit to have a validit conditional chain with the secondary hit chr3_16811234_G_A, for which BETA, SE and MLOG10P are provided both for the original (input) sumstats and the conditioned one. In this case, one can see that the values did not change significantly after conditioning, but it can happen that variants go from being non significant to significant or that betas change considerably. Export to Sheets

• all_logs. This file is simply the collection of all regenie outputs and it's meant for debugging in case something went wrong