Confusion between false positives and read throughs in Table 2
Matt Zhaohui, TCNU
10 June 2015
Actually, the column "False Positives" from Table 2 shows which tools report read throughs (i.e. SOAPfuse, deFuse, FusionCatcher) and which tools do NOT report read throughs (i.e. JAFFA, TopHat-Fusion). Basically, in Table 2, the read throughs are counted as false positives even that in Table 1 they are not. Therefore Table 2 does not show any discovery rate for any of the tools presented as it is claimed in the article (quote: "JAFFA also demonstrated excellent sensitivity and a very low false discovery rate on 250 bp reads simulating a MiSeq dataset (Table 2"). The datasets used in Table 2 contains reads simulated by BEERS tool which the authors acknowlege in the article that it generates read-throughs which are actually fusion genes.
Read throughs are valid/real fusion genes and in the literature are plenty of cases of very well known and oncogenic fusion genes which are read throughs. For example, the famous and very well known fusion gene FGFR3-TACC3 is a read through (the gap between FGFR3 and TACC3 is less than 50k nt ). Please, see here for articles regarding FGFR3-TACC3
Confusion between false positives and read throughs in Table 2
10 June 2015
Actually, the column "False Positives" from Table 2 shows which tools report read throughs (i.e. SOAPfuse, deFuse, FusionCatcher) and which tools do NOT report read throughs (i.e. JAFFA, TopHat-Fusion). Basically, in Table 2, the read throughs are counted as false positives even that in Table 1 they are not. Therefore Table 2 does not show any discovery rate for any of the tools presented as it is claimed in the article (quote: "JAFFA also demonstrated excellent sensitivity and a very low false discovery rate on 250 bp reads simulating a MiSeq dataset (Table 2"). The datasets used in Table 2 contains reads simulated by BEERS tool which the authors acknowlege in the article that it generates read-throughs which are actually fusion genes.
Read throughs are valid/real fusion genes and in the literature are plenty of cases of very well known and oncogenic fusion genes which are read throughs. For example, the famous and very well known fusion gene FGFR3-TACC3 is a read through (the gap between FGFR3 and TACC3 is less than 50k nt ). Please, see here for articles regarding FGFR3-TACC3
http://scholar.google.com/scholar?q=FGFR3-TACC3
Competing interests
No competing interests.