updated readme

readme.md

@@ -38,3 +38,22 @@ bsub -J collate_results -n 4 -q normal -R "select[mem>16000] rusage[mem=16000] s
 #### Overview
 
 ![workflow-overview](overview_figure.png)
+
+#### Summary statistic preparation
+
+- Summary statistics were preprocessed following the criteria outline [here](https://github.com/opentargets/sumstat_data)
+- Allele / effect direction harmonisation scripts can be found [here](https://github.com/opentargets/sumstat_harmoniser)
+
+#### Top loci detection
+
+Independently associated top loci will be detected with GCTA stepwise selection procedure (cojo-slct) using UK10K (ALSPAC + TwinsUK) genotypes (N=3,781) as an LD reference. Index variants (top variant at each locus) will be kept if both the conditional and nominal p-values are less than 1e-5.
+
+Variant in the MHC region were pre-filtered (6:28477797–33448354 GRCh37).
+
+#### Per locus conditional analysis
+
+Where multiple index SNPs are found at the same locus (within 500kb of each other), perform GCTA single-variant association analysis conditional on other index SNPs at the locus.
+
+#### Credible set analysis
+
+Credible set analysis is to be conducted for each associated locus using the above conditional summary statistics. For all variants in a region around the index variant (±500kb), calculate approximate Bayes factors (ABF). ABFs are currently calculated using the `approx.bf.p` [method](https://github.com/chr1swallace/coloc/blob/master/R/claudia.R#L67) re-implemented from the coloc package. The python implementation from our current finemapping pipeline can be seen here. Variants are ordered by their posterior probabilities (PP) and sequentially added to the credible set until the cumulative sum is >0.95 (95% credible set).