Table 1.

Summary of input sequences

	Human					Mouse
	RefSeq	mRNA	EST	RefSeq	mRNA	EST	RefSeq	mRNA	EST
Raw data from GenBank	25,975	133,271	5,426,061	40,568	113,526	3,918,650	21,937	11,779	538,134
No. of aligned sequences onto the genome after initial filtering	25,665	118,034	4,836,878	38,137	101,645	3,467,066	20,867	9996	487,771
No. of sequences after removal of bad alignments^a	24,895 (96%)	112,933 (84%)	4,408,552 (81%)	37,268 (92%)	100,798 (89%)	3,348,841 (85%)	20,759 (94%)	9871 (84%)	471,043 (88%)
No. of spliced sequences^b	22,649 (91%)	86,897 (77%)	2,076,217 (47%)	29,948 (80%)	68,912 (68%)	1,315,511 (39%)	18,289 (88%)	8404 (85%)	169,604 (36%)

↵a Sequences included in the final clustering of ECgene (percentage of aligned sequences)
↵b Input sequences for transcript assembly procedure (percentage of multi-exon sequences out of all sequences in the ECgene)

ECgene: Genome-based EST clustering and gene modeling for alternative splicing