Question about gene fusion

Question

[mangfu100]

hello

I have a question while i am reading a paper that is related to a gene fusion discovery.

below is some of the paper's content.

A paired end read will be a member of multiple valid clusters as a result of homology between genes, exon redundancy between transcripts, and the multiplicity of valid clusterings. Fusion events are rare when compared with the event of detecting a discordant paired end read given the existence of a fusion event that would generate that read. Thus we seek an assignment of each paired end read to a single fusion event (valid cluster) that minimizes the number of fusion events.

(Valid cluster represents a potential fusion event implied by a set of discordant alignments)

i don't understand the bold text. I think the verb "minimizes" should be changed to "maximizes " because

It maybe increase the number of gene fusion by calculating the single paired end read to a single gene fusion.

I hope you understand my basic questions. I'm beginner in this field.

Anyway I am looking forward to your reply.

Thank you.

Answer 1

[Sean Davis]

The problem that the authors are trying to address is the problem of false positive results in the setting of reads that can map to multiple locations. They are suggesting to use a heuristic that takes all the data available and seeks a minimal set of fusions from the set of all potential fusions. So, I believe "minimizes" is what the authors meant.

[mangfu100]

Thank you!

I appreciate a your answer and it gave me a insight to this field.