Loop model for learning sequence patterns of insulator loops

TT Tuan Trieu
AM Alexander Martinez-Fundichely
EK Ekta Khurana
request Request a Protocol
ask Ask a question
Favorite

We built DeepMILO by combining learned features of the anchor and anchor orientation models as shown in (Fig. 3b). The output layer of the anchor model was removed, and the two fully connected layers were replaced by two new layers with the same settings to make a new sub-model. Then, the output from this new sub-model was concatenated with outputs from the anchor and anchor orientation model. The loop model uses sequence features of anchors and predicted outcomes from the anchor model and the anchor orientation model for its prediction. In training, validation, and test sets, positive samples are true insulator loop and negative samples consists of non-loop types 1, 2, 3, 4, and 5 with a ratio of 50%:10%:10%:20%:10%, respectively.

Do you have any questions about this protocol?

Post your question to gather feedback from the community. We will also invite the authors of this article to respond.

0/150

tip Tips for asking effective questions

+ Description

Write a detailed description. Include all information that will help others answer your question including experimental processes, conditions, and relevant images.

post Post a Question
0 Q&A