a Changes in median lifespan in wildtype and hsf-1 overexpressing (hsf-1 OE) worms after RNAi against HSF-1 target genes (sample numbers for all groups can be found within Supplementary Data 1). b, c STRING networks of screened genes that, when knocked down, were found to (b) decrease hsf-1 OE lifespan by > 20% with no effect on wild-type lifespan, or (c) increase hsf-1 OE lifespan by > 20% but decrease wildtype lifespan by the same degree. Black lines represent interaction by co-expression, and purple lines represent evidence of experimental interaction. ubql-1 was included in (c) for reference, and genes without any connections were removed. d Relative expression of ubql-1 mRNA on day 1 of adulthood in wildtype, ubql-1(tm1574), hsf-1 OE, and hsf-1 OE; ubql-1(tm1574) animals grown on OP50. Values were normalised to the geometric mean of the housekeeping genes rpb-2, pmp-3, and cdc-42. Data plotted are the mean of 4 biological replicates +/− SD (wildtype vs ubql-1(tm1574), p = 0.0008; hsf-1OE vs hsf-1OE;ubql-1(tm1574), p < 0.0001; wildtype vs hsf-1OE, p = 0.0005; ubql-1(tm1574) vs hsf-1OE;ubql-1(tm1574), p = 0.8662). e, f Lifespan of (e) wildtype and hsf-1 OE animals on empty vector and ubql-1(RNAi) (wildtype vs ubql-1(RNAi), p = 0.2886; wildtype vs hsf-1OE, p < 0.0001; hsf-1OE vs hsf-1OE;ubql-1(RNAi), p = 0.0002) and (f) wildtype, ubql-1(tm1574), hsf-1 OE, hsf-1 OE;ubql-1(tm1574) animals grown on OP50 (wildtype vs ubql-1(tm1574), p = 0.0002; wildtype vs hsf-1OE, p < 0.0001; hsf-1OE vs hsf-1OE;ubql-1(tm1574), p < 0.0001). Statistical significance was calculated using (d) two-way ANOVA with Fishers LSD and (e, f) Mantel-Cox log-rank test. ns, not significant (p > 0.05), **p < 0.01, ***p < 0.001, ****p < 0.0001. Full statistics for lifespan trials (including n values) can be found in Supplementary Data 2. Source data are provided as a Source Data file.