Table 2:
Performances of SL-quant with various parameters.
| Dataset | Method | Run time | Mapped SL-containing reads | Trans-splice sites | Site is “AG” consensus (%) |
|---|---|---|---|---|---|
| SRR1585277 | SL-quant | 4 minutes 02 seconds | 65,126 | 6,301 | 6,149 (98) |
| SL-quant -p | 5 minutes 14 seconds | 61,451 | 6,539 | 6,402 (98) | |
| SL-quant -s | 2 minutes 45 seconds | 120,542 | 8,770 | 8,254 (94) | |
| SL-quant -s -p | 6 minutes 58 seconds | 114,948 | 8,436 | 7,957 (94) | |
| Tourasse | 4 minutes 45 seconds | 120,710 | 8,932 | 8,260 (92) | |
| modENCODE_4594 | SL-quant | 9 minutes 51 seconds | 146,358 | 8,247 | 8,081 (98) |
| SL-quant -s | 3 minutes 10 seconds | 258,706 | 10,735 | 9,948 (93) | |
| Tourasse | 5 minutes 08 seconds | 259,284 | 11,155 | 9,953 (89) | |
| random | SL-quant | 3 minutes 20 seconds | 53 | 52 | 34 (65) |
| SL-quant -s | 1m23s | 5,757 | 5,692 | 5,612 (99 a) | |
| Tourasse | 2m24s | 8,890 | 8,777 | 5,612 (64) |
a
The very high proportion of “AG” sites for the random dataset is an artifact caused by the fact that the reads were generated from randomly sampling the genome and that all the C. elegans SL sequences end by AG. -p: paired-end mode; -s: sensitive mode.