cansyl / deepred Goto Github PK

You mention MFSPMAP, BPSPMAP, and CCSPMAP features, and your article on DEEPred also talks about SPMAP. I have not been able to find the software that's used to create these features, given protein sequence. Can you point me to the software, or add links to the software in your README?

Hi all,

I was looking at 4_layer_train.py and I noticed that the cost function uses the TF function: tf.nn.sigmoid_cross_entropy_with_logits

However, the logits argument is A3 which is the output of the sigmoid activation (A3 = tf.sigmoid(Z3))

IIUC, the point of the sigmoid_cross_entropy_with_logits functions is that it has a sigmoid so the model shouldn't have any activation function on the last layer. Thus, I think it should be logits=Z3.

Let me know if I am missing something.

The dropbox URL for the FeatureVectors and Annots shows a not found page.

I am a Ph.D. student metabolomic/specialized plant metabolic responses to pathogens.

Your deep learning approach applied to the protein function prediction is very innovative. I'd like to apply it to my plant data to predict the biosynthetic gene cluster.

Does DEEPred work with plants (Brassica spp) data?

IMPORTANT: These plants contain approximately 6 copies of each gene resulting from duplication and therefore a large question of neofunctionalization and subfunctionalization. This is without taking into account the tandem duplication events on the same chromosome.

Thank you for your reply