Enhancing De Novo Transcriptome Assembly by Incorporating Multiple Overlap Sizes

Table 1

Evaluation of assemblies of mouse simulated data with Velvet using different 𝑘-mer and compare with Euler-mix.

𝑘-mer parameter	17	19	21	23	25	27	29	31	33	35	Euler-Mix; (19–35)
Precision(overlap)	99.93%	99.98%	99.99%	100.00%	99.99%	99.99%	99.99%	99.99%	99.98%	99.99%	99.99%
Recall(overlap)	9.97%	74.26%	80.27%	79.39%	77.44%	74.22%	69.53%	60.53%	42.08%	7.95%	85.32%
F-measure(overlap)	18.14%	85.22%	89.05%	88.51%	87.29%	85.20%	82.02%	75.41%	59.24%	14.73%	92.08%
Precision(consistent)	96.59%	81.29%	93.71%	95.01%	94.50%	93.57%	91.86%	89.02%	85.28%	85.05%	96.31%
Recall(consistent)	8.13%	50.28%	60.73%	60.55%	58.51%	55.28%	50.29%	42.22%	27.58%	4.25%	64.71%
F-measure(consistent)	15.00%	62.13%	73.70%	73.96%	72.27%	69.50%	65.00%	57.27%	41.68%	8.09%	77.41%
Number of contigs >= 100 bp	48236	170563	80166	65348	62817	63982	68447	76573	75896	15378	48183
Mean size (bp)	125.28	261.08	582.66	701.67	708.77	663.93	575.11	445.32	306.37	233.43	1001.79
Largest contig	413	3272	11696	27350	45270	81929	81794	27866	14648	8312	81929
N50	121	311	1112	1697	1949	2011	1798	1230	485	250	2783