Table 3.

Evaluation using a RNA-Seq dataset of 145 M paired-end 101-bp reads from the long RNA-seq of IMR90 cell lines from ENCODE Consortium

Method	Original	SEECER	Quake	Coral
Aligned reads (M)	119.0	123.1 (+3.47%)	121.9 (+2.46%)	–
Proper pairs (M)	81.1	85.4 (+5.4%)	83.5 (+2.9%)	–
Zero error reads (M)	76.2	105.3 (+38.2%)	92.4 (+21.3%)	–
Gain	–	0.58	0.32	–
Assembly full	13 148	18 851 (+43.4%)	14 968 (+13.84%)	–
Assembly 80%	61 522	61 178 (−0.6%)	62 231 (+1.2%)	–
Memory (GB)	–	113	60	>130
Time (hours)	–	40.25	3	–

Percentages in brackets denote performance compared with original data.

‘–’ means not applicable.

The evaluation is based on Ensembl 65 annotation.