I am trying to get fusions of Endogenous Retroviral elements (ERVs) and genes/lncRNA. There are some nice tools out there work quite nicely for gene-gene fusions e.g. fusioncatcher etc.
I am searching something which could work well for ERVs and genes,
I already did something by my own, manipulating outputs from some tools, but there was quite low overlap and moreover results were inconsistent among replicates.
I would appreciate if you can please advice me anything on this.
PS: If I can analyze outputs from TopHat2-fusion or STAR outputs in someway where I get more and significant fusions.
I started recently working on fusions.I have been using STAR-fusion and it seems to provide good results,as in confirming the fusions which have a strong evidence of having a fusion related to a specific gene.
I use a cut-off of Junction reads and spanning pairs both to get a significant support.
Hi Ron - what cutoffs do you use for Junction reads and Spanning fragments? Is it referenced anywhere?
Usually I use Junction Reads > 2 and Spanning Pairs > 1 for fusions with good support. Although there could be novel fusions with only Junction Reads or only Spanning Pairs.