Figure 2: Drosophila miRNA phenotypes in viability and external morphology.

(a) Viability defects following ubiquitous expression of SP library under the tubulin-Gal4 driver. Data are displayed as average per cent viability relative to control; at least six independent replicate batches were analysed for each genotype (analysis of variance, post hoc analysis with Tukey–Kramer multiple comparisons test P≤0.01; error bars, s.d). (b) Benchmark comparison of viability phenotypes with Drosophila miRNA null mutants¹³. ‘Confirmed’ indicates same viability phenotype shared with SP and null. False negative (‘False –ve’) indicates miRNAs where null demonstrated viability impaired phenotypes, but miR-SP lines were viable. False positive (‘False +ve’) indicates miRNAs where a phenotype was observed with miR-SP but not in the null animal. ‘miR family’ represents miRNAs for which a similar seed sequence is shared and miR-SP lines display a viability phenotype that is only confirmed by an individual family member. The denominator for this chart was 99; we excluded all lines for which there was no null available, lines that were not tested by Chen et al.¹³, mutants removing entire clusters of multiple miRs, or mutants for which complementation was inconclusive (Supplementary Data 3). (c–h) Eye morphology defects (arrowheads) following inhibition of miR-92 activity. Genotypes: +/tubulin-Gal4 (c), +/Scramble-SP;tubulin-Gal4/Scramble-SP (d,g), +/miR-92bSP;tubulin-Gal4/miR-92bSP (e,h) +/miR-92bSP;tubulin-Gal4/miR-310SP (f). (i–k) Expression of miR-2aSP results in wing vein patterning abnormalities (j, arrowheads) and an overall reduction in wing blade size (j; scale bar, 200 μm). Average wing size area was quantified in triplicate samples (P=0.004; error bars, s.e.m.) from +/Scramble-SP;tubulin-Gal4/Scramble-SP and +/miR-2aSP;tubulin-Gal4/miR-2aSP animals (k).