Page 63 - TD-2-2

Tumor Discovery Enhanced SV analysis in WGS

commonly used SV detection pipelines consist of Manta , users are required to devise their own merging strategies
[8]
DELLY , LUMPY , GRIDSS (GRIDSS2) [11,12] , etc. Due when performing SV calling in practice. To address
[10]
[9]
to the diverse detection methods employed by various this issue, we have developed a user-friendly R package
callers to identify SV, inconsistencies can arise between called structural variants integration and visualization
the outputs generated by different callers. Therefore, (SVIV), (https://github.com/YULEITSINGTAO/SVIV)
researchers often need to merge 2 to 3 callers’ results to and merge the results from different callers to provide
obtain more confident results and perform downstream better visualization of the SVs. This package allows a
analyses. Although studies have been conducted to direct comparison of the results from different callers by
evaluate the performance of various SV callers , there generating tables and graphs to help researchers identify
[13]
is still no consensus on the optimal approach for merging consistent SVs among different callers callers (Figure 1).
the results obtained from different callers. Consequently, Overall, SVIV provides an efficient and reliable solution for

Figure 1. The workflow of SVIV. The SVIV SV analysis workflow consists of three phases. In the first phase, the data is prepared by importing the structure
variation VCF file using the sample map provided by the user, and filtering the mutations. Users can use the available functions in SVIV to combine the
structure mutations and obtain the callers’ combined structure variations. In the second phase, the callers’ combined SVs are separated by mutation type
and aligned into windows provided by the user to collect the SV features. In the third phase, the SVs are visualized using the visualization functions
provided by SVIV.
Abbreviations: SVIV: Structural variants integration and visualization; SV: Structural variation; VCF: Variant call format.

Volume 2 Issue 2 (2023) 2 https://doi.org/10.36922/td.0894