Visualize QC and Results

Plots are a fast way to spot problems. Use them after a pipeline run or after running credtools qc.

Make a Summary QC Plot

For a pipeline locus directory:

credtools plot results/locus_1 \
  --type summary \
  --output results/locus_1/qc_summary.png

For a standalone QC directory:

credtools plot qc_results \
  --type summary \
  --output qc_results/qc_summary.png

Plot One Locus

credtools plot results/locus_1 \
  --type locusplot \
  --output results/locus_1/locusplot.png

This plot helps you check whether high-PIP variants line up with the association signal and LD pattern.

Common Plot Types

Plot type	Use it for
summary	quick QC overview
locus_qc	per-locus QC panels
locusplot	p-values, PIPs, and LD context
lambda_s	summary-statistic and LD consistency
maf_corr	allele-frequency agreement
locus_pvalues	p-value shape inside a locus
zscore_qq	z-score behavior
ld_decay	LD decay by population or cohort
ld_4th_moment	LD structure checks
snp_missingness	which studies are missing variants

Let CREDTOOLS Guess the Plot

If you do not pass --type, CREDTOOLS tries to infer the plot from the input:

credtools plot results/locus_1 --output results/locus_1/auto_plot.png

This is useful for quick checks. Use --type when you want reproducible reports.

Size and Format

credtools plot results/locus_1 \
  --type locusplot \
  --output results/locus_1/locusplot.pdf \
  --width 14 \
  --height 8 \
  --dpi 300

The output extension controls the file type: PNG, PDF, or SVG.

Use plots to choose what to inspect, not as the final evidence

When a plot looks odd, open the matching table and check exact values. The plot is the map; the table is the record.