LogR calculation for tumor-only WES

[berguner]

Hi,
First of all thank you for this great tool.

I have some tumor-only WES samples that I would like to analyse but I am not sure how LogR values would be calculated. Here it says that read counts are normalized for median depth of coverage across autosomes but that wouldn't be ideal. Is it possible to use mean depth of coverage of all the targeted regions in a bunch of normal samples so that coverage bias of the enrichment probes would be taken into account?

[zhangzhhcb]

Hi, I don't understand your meaning about the "mean depth of coverage of all the targeted regions in a bunch of normal samples", considering you said it is tumor-only WES data. Please refer to this post for your question as well. (#151)

[berguner]

I wanted to ask if calculation of LogR values based on a panel of normals was possible. It seems like that was also recommended in #151 but no one has implemented it yet for WES.

[zhangzhhcb]

Hi, the best option is the paired normal. Using a panel of normals may cause noisy results and we haven't evaluated the impact yet. However, if paired normals are not available, a panel of normals is one option.