Integrate structural variations predicted by different methods

Description

Integrate predictions of different tools to provide more reliable structural variations.

Usage

1
2
3
    methodsMerge(..., others=NULL, 
    overLapPerDel = 0.8, overLapPerDup = 0.8, overLapPerInv = 0.8,
    numMethodsSupDel = 2, numMethodsSupDup = 2, numMethodsSupInv = 2)

Arguments

...

results of different SVs predictions read in to R by intansv.

others

a data frame of structural variations predicted by other tools.

overLapPerDel

Deletions predicted by different methods that have reciprocal coordinate overlap larger than this threshold would be clustered together

overLapPerDup

Duplications predicted by different methods that have reciprocal coordinate overlap larger than this threshold would be clustered together

overLapPerInv

Inversions predicted by different methods that have reciprocal coordinate overlap larger than this threshold would be clustered together

numMethodsSupDel

Deletion clusters supportted by no more than this threshold of read support would be discarded

numMethodsSupDup

Duplication clusters supportted by no more than this threshold of read support would be discarded

numMethodsSupInv

Inversion clusters supportted by no more than this threshold of read support would be discarded

Details

A structural variation (deletion, duplication, inversion et al.) may be reported by different tools. However, the boundaries of this structural variation predicted by different tools don't always agree with each other. Predictions of different methods with reciprocal overlap more than 80 percent were merged. Structural varions supported by only one method were discarded.

Value

A list with the following components:

del

the integrated deletions of different methods.

dup

the integrated duplications of different methods.

inv

the integrated inversions of different methods.

Author(s)

Wen Yao

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
    breakdancer <- readBreakDancer(system.file("extdata/ZS97.breakdancer.sv",
                                   package="intansv"))
    str(breakdancer)

    cnvnator <- readCnvnator(system.file("extdata/cnvnator",package="intansv"))
    str(cnvnator)

    svseq <- readSvseq(system.file("extdata/svseq2",package="intansv"))
    str(svseq)

    delly <- readDelly(system.file("extdata/delly",package="intansv"))
    str(delly)

    pindel <- readPindel(system.file("extdata/pindel",package="intansv"))
    str(pindel)

    sv_all_methods <- methodsMerge(breakdancer,pindel,cnvnator,delly,svseq)
    str(sv_all_methods)
    
    sv_all_methods.1 <- methodsMerge(breakdancer,pindel,cnvnator,delly,svseq,
                                 overLapPerDel=0.7)
    str(sv_all_methods.1)

    sv_all_methods.2 <- methodsMerge(breakdancer,pindel,cnvnator,delly,svseq,
                                 overLapPerDel=0.8, numMethodsSupDel=3)
    str(sv_all_methods.2)