Giter VIP home page Giter VIP logo

Comments (12)

nadavbra avatar nadavbra commented on July 24, 2024

Could you please clarify your question? I'm not sure what you mean by "the global feature extracted by ProteinBERT" and "the original label".

from protein_bert.

xixinhy avatar xixinhy commented on July 24, 2024

Wasn't the initial input a sequence and a label? I want to know if the extracted global feature vector corresponds one-to-one to the original input label, that is, if the extracted global feature vector corresponds to the order of the original sequence?

from protein_bert.

nadavbra avatar nadavbra commented on July 24, 2024

If you refer to the 8943-dimensional vector of GO annotations - then yes, it's provided both as input and as output of the same GO annotation at each position. You may find the discussion in issue #6 relevant.

from protein_bert.

xixinhy avatar xixinhy commented on July 24, 2024

I entered a biomolecular structure instead of a Go comment. I want to get the features of the sequence. is the feature I need to extract a local feature instead of a global one?

from protein_bert.

nadavbra avatar nadavbra commented on July 24, 2024

I don't understand your question. Could you please elaborate on what you did / want to do?

from protein_bert.

xixinhy avatar xixinhy commented on July 24, 2024

I'm sorry, maybe my expression isn't perfect, I want to use this model of yours, enter my biomolecular structure, to get the feature vector, but I'm not sure if I should use local or global features, And another problem is, do I want to know if the order of the features I get is the same as the order of the Biomolecular Structure I entered?

from protein_bert.

nadavbra avatar nadavbra commented on July 24, 2024

If you want to use the pre-trained model, it only takes protein sequence and (optionally) GO annotations as input. If you want to use structure input, you should train a model that accepts that as input (optionally also with features/outputs from ProteinBERT).

from protein_bert.

xixinhy avatar xixinhy commented on July 24, 2024

Exactly,I want to use the pre-trained model, and use the protein sequence as input. I want to extract the feature vector of the protein sequence, so which feature vector do I choose, the global representations or the local representations?

from protein_bert.

nadavbra avatar nadavbra commented on July 24, 2024

So you want to train a model that takes a protein sequence as input and outputs the structure of that protein?

from protein_bert.

xixinhy avatar xixinhy commented on July 24, 2024

No, no ,I want the eigenvectors of that protein, but I don't know if it's global or local. What do you suggest?

from protein_bert.

nadavbra avatar nadavbra commented on July 24, 2024

If you want me to be able to help you, you will have to be clearer about what you're hoping to achieve.

from protein_bert.

xixinhy avatar xixinhy commented on July 24, 2024

My goal is that I want to do the classification task of protein sequences, but the number of my protein is only more than 400

from protein_bert.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.