| Protein . | Ratio proximal/middle . | Ratio distal/middle . |
|---|---|---|
| eIF4AIII | 9.37 (8.39–10.50) | 2.85 (2.48–3.27) |
| Y14 | 8.53 (7.63–9.54) | 2.87 (2.49–3.30) |
| Mago | 7.90 (7.09–8.79) | 2.63 (2.30–3.01) |
| Btz | 4.04 (3.48–4.69) | 2.65 (2.25–3.13) |
| Protein . | Ratio proximal/middle . | Ratio distal/middle . |
|---|---|---|
| eIF4AIII | 9.37 (8.39–10.50) | 2.85 (2.48–3.27) |
| Y14 | 8.53 (7.63–9.54) | 2.87 (2.49–3.30) |
| Mago | 7.90 (7.09–8.79) | 2.63 (2.30–3.01) |
| Btz | 4.04 (3.48–4.69) | 2.65 (2.25–3.13) |
Gold particle counts, pooled over four experiments, were analyed using a generalized linear model with Poisson distribution and a log link. Offsets were used and correspond to the estimated segment proportion in the experiment (∼20% proximal, 60% middle, and 20% distal). The table shows gold particles/length unit ratios for the proximal and distal segments against the middle segment, together with 95% confidence intervals. The individual measurements are presented in Table S3. The difference in gold particles/length unit ratios between proteins (eIF4AIII, Y14, and Mago compared with Btz) is highly significant (likelihood ratio test, P < 0.00005) and is caused by the lower gold particles/length unit ratio for Btz in the proximal/middle segment.