Table 1

Overall descriptions of SAGE librariesa

LibrarybD1D2CL1CL2CL3CL4CL5CL6I1I2Total
Tags32157324201046724879379262322231864323003394235743294920
Unique tagsc852968724127781511711782486358286113551150486658
Genesd841168454112778311660778885838228112971145077746
>500
Unique tagsc0001 (5)001 (2)1 (5)1 (5)1 (7)5 (3)
Genesd00010011115
>50 and ≤500
Unique tagsc69 (29)71 (31)14 (11)35 (13)76 (21)46 (17)65 (23)62 (22)38 (12)38 (11)514 (20)
Genesd68701435764665623838512
>3 and ≤50
Unique tagsc1112 (35)1196 (39)345 (38)1033 (41)1389 (37)818 (39)1104 (37)1075 (37)1211 (38)1376 (39)10659 (38)
Genesd11111193344102813867817109510601203136810605
≤3
Unique tagsc7348 (36)5605 (30)3768 (51)6746 (41)10246 (42)6960 (45)7465 (38)7148 (35)10105 (45)10089 (43)75480 (39)
Genesd72195572375467196799692574227105100551004374832
  • a The libraries are: H126 and HX, human pancreas duct epithelium from short-term cultures; Hs766T, Panc-1, CAPAN1, CAPAN2, PL45, and AsPc1, pancreatic cancer cell lines; 91-16113 and 96-6252, primary pancreatic adenocarcinomas.

  • b D1 and D2 are normal pancreatic ductal epithelium, HX and H126, respectively. CL1 to CL6 are pancreatic cancer cell lines, Hs766T, Panc-1, CAPAN1, CAPAN2, PL45, and AsPcl. I1 and I2 are primary pancreatic cancer, 91-16113 and 96-6252, respectively.

  • c Number of different transcripts represented by the total tags analyzed. Numbers in parentheses indicates the mass fraction represented by the indicated abundance class (the sum of tag counts in a category divided by total tags for the sample, in percentage).

  • d Number of genes, corrected by the estimated SAGE sequence error rate (6.8%) (14) . These numbers are obtained by elimination of ESTs, mitochondrial DNA sequences, and repetitive DNA sequences from a dataset of total genes that matched to an entry of SAGEtag to UniGene Mapping database (http://ncbi.nlm.nih.gov/pub/sage/map/). A database, SAGEmap_tag_ug.rel.NlaIII.hs, representing those tag to gene assignments deemed reliable by CGAP, was used.