version: "1.0.1" name: bio-causal-genomics-fine-mapping description: Resolves GWAS associations to candidate causal variants and credible sets via SuSiE, susie_rss, FINEMAP, CAVIAR, DAP-G, PAINTOR, PolyFun, SuSiEx, MultiSuSiE, and FOCUS. Use when narrowing a GWAS lead SNP to a 95 percent credible set, choosing between in-sample and reference LD, calibrating non-sparse loci with SuSiE-inf or FINEMAP-inf, integrating functional priors via PolyFun, fine-mapping across ancestries with SuSiEx, diagnosing LD mismatch via estimate_s_rss and kriging_rss, handling HLA or long-range LD, or feeding credible sets into coloc.susie for colocalization. tool_type: r primary_tool: susieR

Version Compatibility

Reference examples tested with: susieR 0.12.27+, coloc 5.2.3+, FINEMAP 1.4.2+, PolyFun (head of omerwe/polyfun 2024), PAINTOR V3.0, SuSiEx (head of getian107/SuSiEx), DAP-G (head of xqwen/dap), pyfocus 0.8+, R 4.3+, PLINK 1.9 / 2.0.

Before using code patterns, verify installed versions match. If versions differ:

• R: packageVersion('susieR') then ?susie_rss to confirm argument names (e.g., prior_weights vs prior_variance semantics)
• CLI: finemap --help, SuSiEx --help, PAINTOR --help, dap-g --help to confirm flags
• Python: polyfun.py --help

If a call throws an error about an argument that no longer exists, introspect the installed function and adapt rather than retrying.

Fine-Mapping

"Narrow my GWAS locus to the variants likely to be causal" -> Fit a sparse Bayesian regression that propagates LD into posterior inclusion probabilities (PIPs) and credible sets, then validate that credible sets correspond to physically reasonable haplotypes given the LD reference.

• R (summary statistics + LD): susieR::susie_rss(z, R, n, L=10) + estimate_s_rss LD diagnostic
• R (individual-level genotypes): susieR::susie(X, y, L=10)
• CLI (shotgun stochastic search): finemap --sss --in-files master.z --n-causal-snps 5 --prob-tol 0.001
• CLI (cross-ancestry joint): SuSiEx --sst_file=eur.sst,eas.sst --n_gwas=N1,N2 --ref_file=eur.bim,eas.bim --ld_file=eur_ld,eas_ld --chr_col=1,1 --snp_col=2,2 --bp_col=3,3 --a1_col=4,4 --a2_col=5,5 --eff_col=6,6 --se_col=7,7 --pval_col=8,8 --chr=<chr> --bp=<start,end> --out_dir=<dir> --out_name=<name> (column-number flags and --ld_file are required; populations are assigned by the ORDER of the comma-separated --sst_file/--n_gwas/--ref_file/--ld_file lists, not a --pop flag; see SuSiEx --help)
• Python (functional priors): polyfun.py --compute-h2-L2 -> per-SNP priors -> susie_rss with prior_weights=
• Python (TWAS fine-mapping): focus finemap on gene-level Z-scores

Fine-mapping is a Bayesian model selection problem; LD is not noise but structured prior information. Most failure modes trace back to one of three issues: (a) LD reference mismatched to the GWAS sample; (b) the sparse-effects prior being wrong for the locus (polygenic background); or (c) too small an L cap. The estimate_s_rss() lambda and kriging_rss() per-SNP diagnostic catch (a) before downstream credible sets are reported.

Algorithmic Taxonomy

Methodology evolves; verify the latest susieR vignette and the SuSiE-inf paper before locking on a single method. Wang Lab maintains susieR; the IBSS algorithm is stable but argument semantics (e.g., prior_weights vs prior_variance) have changed across versions.

Decision Tree by Experimental Scenario

Critical LD Diagnostic Block (susie_rss)

Goal: Detect LD reference mismatch before reporting credible sets.

Approach: estimate_s_rss() quantifies the global Z-score / LD inconsistency as a scalar; kriging_rss() identifies individual SNPs whose Z-scores are inconsistent with the LD reference (typically genotyping errors, strand flips, or wrong reference panel).

library(susieR)
s_hat <- estimate_s_rss(z = z_scores, R = ld_matrix, n = N)
# s_hat is the inferred scale of LD inconsistency.
# Source: susieR vignette "Diagnostic for summary statistic"; Zou 2022 PLoS Genet.
# Rule of thumb: s_hat < 0.05 acceptable; 0.05-0.10 marginal; > 0.10 refit or change LD reference.
 
cond_z <- kriging_rss(z = z_scores, R = ld_matrix, n = N)
# cond_z$conditional_dist returns per-SNP expected vs observed z; flag |z_obs - z_exp| > 3
# Common cause: strand flip, allele coding mismatch, or single-SNP imputation error.
 
# If diagnostic fails: refit with explicit scale parameter to absorb LD mismatch
fit <- susie_rss(z = z_scores, R = ld_matrix, n = N, L = 10, estimate_residual_variance = TRUE)

Skipping this block is the dominant cause of irreproducible fine-mapping. Always run before reporting credible sets.

Per-Tool Failure Modes

LD reference mismatch (most common)

Trigger: External LD matrix from 1000 Genomes / UK Biobank reference used for a GWAS conducted on a different cohort or ancestry mix.

Mechanism: Z-scores reflect the GWAS sample's LD; the reference R does not. The susie_rss likelihood depends on z' R^{-1} z being consistent with the modeled effects, and inconsistency manifests as spurious credible sets containing tag SNPs from the reference but not from the discovery cohort.

Symptom: estimate_s_rss() lambda > 0.05; kriging_rss() flags many SNPs with |z_obs - z_exp| > 3; credible sets contain physically distant SNPs (anti-correlated in LD with the lead) or include all SNPs at the locus.

Fix: Use in-sample LD whenever the cohort genotypes are accessible (compute with plink --r2 square on the GWAS samples themselves). When only summary statistics are available, ancestry-stratify the LD reference exactly (e.g., 1000G EUR FIN+CEU+GBR+IBS+TSI for a Northern European GWAS, not full EUR). For mixed-ancestry GWAS, fine-map per ancestry then meta-analyze, or move to SuSiEx.

Non-sparse architecture (biobank scale)

Trigger: Locus with one strong signal plus hundreds of weakly associated SNPs (polygenic shoulder); typical at biobank scale.

Mechanism: Vanilla SuSiE assumes a sparse sum-of-single-effects prior. With polygenic background, the model misallocates effects, producing inflated credible sets or many small spurious ones. Cui 2024 (Nat Genet 56:162) showed PIPs from SuSiE in this regime are systematically miscalibrated.

Symptom: Many small credible sets (5-15 per locus); replication in independent cohorts fails for non-lead credible sets; PIP distribution has a heavy tail.

Fix: Use SuSiE-inf or FINEMAP-inf (Cui 2024). These augment the sum-of-single-effects with an infinitesimal random-effect component that absorbs polygenic background. Source: github.com/FinucaneLab/fine-mapping-inf.

L too small

Trigger: Locus with > 5 independent signals (HLA region, APOC1/APOE, LPA, IL6R region for some traits).

Mechanism: SuSiE assumes at most L independent effects. When the true number exceeds L, some signals are absorbed into existing components, distorting PIPs and credible sets for the captured signals.

Symptom: length(fit$sets$cs) equals L (all L slots used); credible set purity for higher-indexed sets is low (fit$sets$purity[,'min.abs.corr'] < 0.5); fits with larger L change top-PIP variants.

Fix: Increase L iteratively (L=10 -> 20 -> 30) until length(fit$sets$cs) < L (susieR auto-prunes unsupported effects so the returned CS count is the effective L). For HLA, start at L=30. The cost is mostly computational, not statistical: SuSiE prunes unused slots, so L=30 is safe when L=10 was right.

prior_weights vs prior_variance confusion (PolyFun integration)

Trigger: Passing PolyFun output to susie_rss with prior_variance=polyfun_priors (wrong argument).

Mechanism: prior_variance in susie_rss is a single scalar (or vector of length L) for the per-effect variance, NOT a per-SNP probability. prior_weights is the per-SNP causal probability vector (sums to ~1). Passing PolyFun's per-SNP prior to prior_variance is silently accepted but applies a numerically nonsensical per-effect variance.

Symptom: PIPs nearly identical to the uniform-prior fit; functional annotations appear to have no effect.

Fix: Use prior_weights = polyfun_priors$SNPVAR (the PolyFun output column is uppercase SNPVAR; R is case-sensitive). Verify with ?susie_rss in the installed version. Reference: github.com/omerwe/polyfun README, Weissbrod 2020 supplementary methods.

Credible-set misinterpretation

Trigger: Reporting per-variant PIP without distinguishing "in credible set" from "high PIP".

Mechanism: The 95 percent credible-set guarantee is P(causal variant in set) >= 0.95. Per-variant PIPs within a set do not necessarily sum to 1 across all variants, and PIPs across overlapping sets can double-count posterior mass.

Symptom: Reporting "the top PIP variant" when the credible set is wide (size > 50); claiming a single variant is causal when the set contains 30 high-LD SNPs.

Fix: Always report (a) number of credible sets, (b) size of each set, (c) purity (fit$sets$purity[,'min.abs.corr']), (d) the top PIP variant within the set as the candidate lead. The credible set is the unit of inference; the top PIP variant is a candidate, not a conclusion.

Cross-ancestry with single-ancestry LD

Trigger: Multi-ancestry meta-analyzed GWAS, then susie_rss with EUR LD.

Mechanism: Meta-analysis z-scores reflect a weighted mix of population LD structures; no single-population LD matrix matches.

Fix: Move to SuSiEx (joint cross-ancestry SuSiE; Yuan 2024). Per-ancestry fine-mapping followed by manual merging loses the shared-causal-variant information that SuSiEx exploits.

Case-control GWAS passing Ntotal instead of Neff

Trigger: Passing n = N_total to susie_rss() for case-control GWAS derived from logistic regression.

Mechanism: susie_rss expects the effective sample size that determined the standard errors. For case-control logistic regression, Neff = 4 / (1/Ncase + 1/Ncontrol); when cases are rare, total N can exceed Neff by 25x or more. Passing Ntotal rescales z-scores into a regime SuSiE never sees and makes the implied prior variance wrong.

Symptom: PIPs systematically biased; credible sets either too narrow (PIPs collapse to a single SNP that is not robust) or too wide (PIPs flatten); replication poor; sometimes z-score scale warnings from susieR.

Fix: Neff = 4 / (1/Ncase + 1/Ncontrol). Example: Ncase=5000, Ncontrol=495000 -> Neff ~= 19,800 (NOT 500,000). For quantitative traits from linear regression, n = N_total is correct. Reference: Privé F et al 2022 HGG Adv 3:100136 (bigsnpr documents Neff handling); Willer 2010 Bioinformatics (METAL Neff convention).

Allele Harmonization with the LD Reference

Trigger: Effect allele in GWAS sumstats differs from the coding/A1 allele in the LD reference panel; or palindromic SNPs (A/T, C/G) carried without strand resolution.

Mechanism: susie_rss treats z and R as defined on the same allele coding. If the effect allele is swapped relative to the LD-reference A1, the sign of z is wrong and the LD row/column for that SNP is implicitly flipped. SNPs matching by rsID can silently swap alleles between sumstats and reference, breaking the z' R z consistency the model relies on.

Symptom: estimate_s_rss lambda inflated despite ancestry-matched panel; kriging_rss flags many SNPs with |z_obs - z_exp| > 3 clustered at SNPs where reference A1 != GWAS effect allele; credible sets pick up tag-only SNPs anti-correlated with the lead.

Fix: Harmonize before fitting:

harmonize_z_to_ref <- function(z, gwas_a1, gwas_a2, ref_a1, ref_a2) {
    palindromic <- (gwas_a1 == 'A' & gwas_a2 == 'T') | (gwas_a1 == 'T' & gwas_a2 == 'A') |
                   (gwas_a1 == 'C' & gwas_a2 == 'G') | (gwas_a1 == 'G' & gwas_a2 == 'C')
    flip <- (gwas_a1 == ref_a2) & (gwas_a2 == ref_a1)
    z[flip] <- -z[flip]
    drop <- palindromic | !((gwas_a1 == ref_a1 & gwas_a2 == ref_a2) | flip)
    list(z = z, keep = !drop)
}

Drop palindromic SNPs at MAF > 0.42 (ambiguous strand); or resolve via external strand info (TopMed, 1000G strand files). TwoSampleMR::harmonise_data() offers an alternative implementation. See causal-genomics/colocalization-analysis for an equivalent harmonize helper used downstream.

Reconciliation: When Methods Disagree

Operational rule: For high-confidence reporting, require that (a) estimate_s_rss() lambda < 0.05; (b) at least one credible set has purity > 0.5 (min_abs_corr >= 0.5, equivalent to r2 >= 0.25); (c) the lead PIP variant within that set is reproduced by an independent method (FINEMAP, SuSiEx, or in-sample SuSiE if reference-LD was used). Anything failing these three is exploratory.

Quantitative Thresholds

Functional Priors with PolyFun

Goal: Use genome-wide stratified LDSC heritability to weight per-SNP causal priors, sharpening PIPs at coding, conserved, and regulatory variants.

Approach: Run PolyFun once genome-wide to estimate per-SNP h2 from the baseline-LF annotation set; extract per-SNP causal prior; pass to susie_rss as prior_weights.

# Parametric route: L2-regularized S-LDSC writes per-SNP priors directly (--no-partitions)
polyfun.py --compute-h2-L2 --no-partitions \
    --output-prefix polyfun_h2 \
    --sumstats gwas_munged.sumstats \
    --ref-ld-chr UKB_baseline_LF/baselineLF2.2.UKB. \
    --w-ld-chr UKB_baseline_LF/weights.UKB.
# Per-SNP priors written to polyfun_h2.<CHR>.snpvar_ridge_constrained.gz
 
# Non-parametric route (finer, optional): drop --no-partitions above, then add an
# intermediate LD-score step before re-estimating binned per-SNP h2:
#   polyfun.py --compute-ldscores --output-prefix polyfun_h2 ...
#   polyfun.py --compute-h2-bins --output-prefix polyfun_h2 --sumstats gwas_munged.sumstats --w-ld-chr UKB_baseline_LF/weights.UKB.

library(susieR)
priors <- read.table('polyfun_h2.6.snpvar_ridge_constrained.gz', header = TRUE)
priors <- priors[match(gwas_df$SNP, priors$SNP), ]
prior_w <- priors$SNPVAR / sum(priors$SNPVAR, na.rm = TRUE)
 
fit <- susie_rss(z = z_scores, R = ld_matrix, n = N, L = 10,
                 prior_weights = prior_w)

UKB baseline-LF priors are pre-computed EUR-only at data.broadinstitute.org/alkesgroup/UKBB_LD/ for hg19 and hg38. For EAS, AFR, or SAS GWAS, the EUR weights are NOT valid: functional-prior fine-mapping in a non-EUR ancestry requires baseline-LF annotations matched to that ancestry. For ancestries lacking matched baseline-LF (admixed, under-represented), accept reduced power and run uniform-prior susie_rss; applying EUR weights to non-EUR sumstats produces miscalibrated PIPs that look sharper than reality.

Manual Coding-Variant Priors Without PolyFun

For postdocs without PolyFun infrastructure or with single-locus inputs, manual annotation-based priors are a reasonable approximation (Hutchinson 2020 Hum Mol Genet 29:R81). As a stated convention, coding variants get ~10x uniform weight; broadly conserved variants ~5x (binned by CADD-PHRED quantile).

build_manual_priors <- function(vep_df, cadd) {
    w <- rep(1, nrow(vep_df))
    w[vep_df$Consequence %in% c('missense_variant', 'stop_gained', 'splice_donor_variant',
                                'splice_acceptor_variant', 'frameshift_variant')] <- 10
    w[cadd >= quantile(cadd, 0.95, na.rm = TRUE)] <- pmax(w[cadd >= quantile(cadd, 0.95, na.rm = TRUE)], 5)
    w / sum(w)
}
fit <- susie_rss(z = z_scores, R = ld_matrix, n = Neff, L = 10, prior_weights = build_manual_priors(vep, cadd))

Report the prior construction explicitly; reviewers will ask whether the prior was tuned post hoc.

Cross-Ancestry Fine-Mapping with SuSiEx

Goal: Jointly fine-map a locus across multiple ancestries assuming shared causal variants but population-specific LD.

Approach: Per-ancestry summary statistics + per-ancestry LD reference; SuSiEx runs a joint SuSiE model with population-specific R matrices. SuSiEx assigns populations by the ORDER of the comma-separated --sst_file/--n_gwas/--ref_file/--ld_file lists (there is no --pop flag); keep all four lists in the same population order.

SuSiEx \
    --sst_file=eur_sumstats.txt,eas_sumstats.txt,afr_sumstats.txt \
    --n_gwas=500000,200000,80000 \
    --ref_file=1000G_EUR,1000G_EAS,1000G_AFR \
    --ld_file=eur_ld,eas_ld,afr_ld \
    --out_dir=susiex_out \
    --out_name=locus1 \
    --chr=6 --bp=30000000,31000000 \
    --chr_col=1,1,1 --snp_col=2,2,2 --bp_col=3,3,3 \
    --a1_col=4,4,4 --a2_col=5,5,5 --eff_col=6,6,6 \
    --se_col=7,7,7 --pval_col=8,8,8 \
    --level=0.95

The output includes per-population PIPs and a joint credible set. Credible sets from SuSiEx are typically 2-5x smaller than EUR-only susie_rss when AFR is included, because AFR shorter LD blocks resolve EUR-tagged regions.

FINEMAP CLI Pattern

Goal: Independent confirmation via shotgun stochastic search.

Approach: Build .z, .ld, and master files; run FINEMAP with --sss and parse the .snp and .cred outputs.

# .z file format: snp chromosome position allele1 allele2 maf beta se
# .ld file: square LD matrix, space-separated, no header
 
cat > locus.master <<'EOF'
z;ld;snp;config;cred;log;n_samples
locus.z;locus.ld;locus.snp;locus.config;locus.cred;locus.log;500000
EOF
 
finemap --sss \
    --in-files locus.master \
    --n-causal-snps 5 \
    --prob-tol 0.001 \
    --n-iterations 100000 \
    --n-convergence 5000
 
# Parse:
# locus.snp -> per-variant prob (PIP), log10bf
# locus.cred -> credible sets at increasing causal counts
# locus.config -> top configurations

FINEMAP and SuSiE agree when sparsity holds; disagreement often reveals non-sparse loci that need SuSiE-inf.

Coloc.susie Integration

Goal: Test colocalization between two traits using credible sets, not single SNPs.

Approach: Fit susie_rss separately per trait; pass both susie objects to coloc.susie; per-credible-set colocalization probabilities are returned.

library(coloc)
 
fit_trait1 <- susie_rss(z = z1, R = ld_matrix, n = N1, L = 10)
fit_trait2 <- susie_rss(z = z2, R = ld_matrix, n = N2, L = 10)
 
coloc_res <- coloc.susie(fit_trait1, fit_trait2)
# coloc_res$summary: per-credible-set PP.H4 (shared causal probability)
print(coloc_res$summary)

PP.H4 > 0.8 per credible set is the conventional shared-causal threshold; weaker thresholds suggest distinct or conditional signals. See causal-genomics/colocalization-analysis.

HLA and Long-Range LD: When to Stop

The HLA region (chr6:28-34 Mb), chromosome 8 inversion (chr8:8-12 Mb), and a handful of other extended LD blocks violate the assumptions of every fine-mapping method.

Symptoms of irrecoverable LD structure: Credible sets contain 30+ SNPs at low purity even with L=30; SuSiE-inf credible sets remain wide; kriging_rss flags hundreds of SNPs.

Options:

• Stratify by classical HLA allele (HIBAG, SNP2HLA imputation) and test allelic series
• Conditional analysis on the lead variant before fine-mapping the residual
• Exclude the region from genome-wide fine-mapping summaries and report separately
• For chr8 inversion: stratify by inversion genotype if known

Document the caveat in any methods section; standard PIPs at HLA are not interpretable as causality estimates.

TWAS Fine-Mapping (FOCUS) -- delegated

Variant-level fine-mapping cannot distinguish causal genes among co-regulated TWAS hits. FOCUS / MA-FOCUS extend fine-mapping to the predicted-expression level; see causal-genomics/transcriptome-wide-association for the full FOCUS workflow and reconciliation with variant-level credible sets.

Required Reporting Schema for Fine-Mapping

Every locus reported should carry these columns; missing fields are the most common reviewer complaint.

Anticipated Reviewer Pushback

Common Errors

References

• Wang G, Sarkar A, Carbonetto P, Stephens M 2020 J R Stat Soc B 82:1273 (SuSiE / IBSS)
• Zou Y, Carbonetto P, Wang G, Stephens M 2022 PLoS Genet 18:e1010299 (susie_rss for summary statistics)
• Cui R, Elzur RA, Kanai M, Ulirsch JC, Weissbrod O et al 2024 Nat Genet 56:162 (SuSiE-inf / FINEMAP-inf for non-sparse loci)
• Benner C, Spencer CC, Havulinna AS, Salomaa V, Ripatti S, Pirinen M 2016 Bioinformatics 32:1493 (FINEMAP)
• Hormozdiari F, Kostem E, Kang EY, Pasaniuc B, Eskin E 2014 Genetics 198:497 (CAVIAR)
• Wen X, Lee Y, Luca F, Pique-Regi R 2016 AJHG 98:1114 (DAP-G)
• Kichaev G, Yang WY, Lindstrom S, Hormozdiari F, Eskin E et al 2014 PLoS Genet 10:e1004722 (PAINTOR)
• Weissbrod O, Hormozdiari F, Benner C, Cui R, Ulirsch J et al 2020 Nat Genet 52:1355 (PolyFun + functional priors)
• Yuan K, Longchamps RJ, Pardinas AF, Yu M, Chen TT et al 2024 Nat Genet 56:1841 (SuSiEx cross-ancestry)
• Rossen J, Shi H, Strober BJ, Zhang MJ, Kanai M, McCaw ZR, Liang L, Weissbrod O, Price AL 2025 Nat Genet 58:67 (MultiSuSiE; doi:10.1038/s41588-025-02450-5)
• Mancuso N, Freund MK, Johnson R, Shi H, Kichaev G et al 2019 Nat Genet 51:675 (FOCUS for TWAS fine-mapping)
• Wallace C 2021 PLoS Genet 17:e1009440 (coloc.susie integration)
• Schaid DJ, Chen W, Larson NB 2018 Nat Rev Genet 19:491 (fine-mapping review)
• Hutchinson A, Asimit J, Wallace C 2020 Hum Mol Genet 29:R81 (fine-mapping review)

Related Skills

• causal-genomics/colocalization-analysis - coloc.susie consumes susie_rss credible sets; equivalent harmonize helper
• causal-genomics/effector-gene-prioritization - Downstream gene-assignment from credible-set variants
• causal-genomics/transcriptome-wide-association - FOCUS / MA-FOCUS for gene-level fine-mapping
• causal-genomics/genomic-sem - Joint multi-trait fine-mapping when credible sets are shared across traits
• causal-genomics/mendelian-randomization - Fine-mapped variants as cis-instruments
• causal-genomics/pleiotropy-detection - Per-credible-set pleiotropy testing
• atac-seq/enhancer-gene-linking - ABC / ENCODE-rE2G linking credible-set variants to target genes
• population-genetics/linkage-disequilibrium - Constructing LD matrices for susie_rss
• population-genetics/association-testing - Upstream GWAS summary statistic generation
• workflows/gwas-pipeline - End-to-end GWAS pipeline producing fine-mapping input
• variant-calling/variant-annotation - Annotating credible-set variants with VEP / coding consequence
• pathway-analysis/go-enrichment - Downstream gene-level interpretation of credible-set targets