DiscoveryDNA / team_neural_network

Neural Network of enhancer regions
13 stars 7 forks source link

Incorporating TFBS scores from negative strand #34

Open iamciera opened 5 years ago

iamciera commented 5 years ago

At this point we are isolating just the positive strand during TFBS subsetting. It would be great if we could incorporate the negative strand scores of TFBS. This is the most important data to add.

There were two options we discussed for this.

  1. Which was brought up by @thethomaslane, was to just run the complement of the TFBS motif. But under the current architecture, this would mean that the original and "complement" TFBS score would be different. Or would it? Anyway this would mean that we only need one nucleotide sequence.

  2. Then @adfost mentioned another way, but I am not sure what you meant. @adfost, can you briefly articulate what you meant?

Either way we do it, I think it needs to be inputed soon. Any other ideas?