search

tanlongzhi / dip-c

Tools to analyze Dip-C (or other 3C/Hi-C) data
61 stars 18 forks source link

Just thinking out loud #22

Closed tarak77 closed 5 years ago

tarak77 commented 5 years ago

Hey @tanlongzhi , At present the pipeline requires the SNP VCF files to be given, but i was just curious whether using googles DeepVariant (a deep learning model for finding the variants) be useful or even help us find more phased legs?? I was thinking of implementing it. Let me know it makes sense to use it..

tanlongzhi commented 5 years ago

Hey @tarak77,

If you're talking about generating a phased SNP file from Dip-C data (instead of requiring a phased SNP file from external sources), yes I believe that's possible given a certain number of single cells. I'm thinking about implementing this in the future, but currently busy with a different project.

In bulk Hi-C, I think the first paper on this topic is Selvaraj et al. 2013; a more recent paper is Edge et al. 2016.

If you're talking about increasing the percentage of phased leg given a phased SNP file, I don't think much improvement can be made, because this percentage is determined simply by how much the mother differs from the father.