Extended Data Fig. 5: Variant co-occurrence heatmap.

Heatmap displaying the observed co-occurrences in ALS patients among variant pairs. For genes identified in the URV burden analysis, an individual is considered a carrier if they had at least one qualifying variant in that gene. The lower triangle shows the log-transformed odds ratio, where red indicates a synergistic effect (greater co-occurrence than expected under an additive risk model), and blue indicates an antagonistic effect (lesser co-occurrence than expected under an additive risk model). The upper triangle shows the statistical significance from a permutation analysis (−log₁₀(P-value)), with darker green indicating a more significant P-value. Pairs for which there was no co-occurrence in the full dataset (cases and controls) are greyed out. In addition to variants identified in this study, we included C9orf72 repeat expansion status in these co-occurrence analyses, which was available for 8,610 (66%) of cases. P-values are two-tailed and presented uncorrected for multiple testing.