{ "topic": { "id": 402597, "title": "Bioinformatics on Kaggle. Biopython.", "authorName": "Marília Prata", "commentCount": 12, "votes": 46, "postDate": "2023-04-19T01:14:13.113000" }, "comments": [ { "id": 2226788, "authorName": "Alexander Chervov", "votes": 4, "postDate": "2023-04-19T08:05:18.200000", "content": "

Thanks for kind words !

\n

It is a regular challenge in bioinformatics and huge literature exists:

\n

https://scholar.google.fr/scholar?start=0&q=cafa+protein+function+prediction&hl=en&as_sdt=0,5&as_vis=1

\n

ChatGPT:

\n

Question: best approaches to solve \"CAFA challenge - Critical Assessment of Functional Annotation \"

\n

Answer:
\nThe CAFA (Critical Assessment of Function Annotation) challenge is a community-wide effort to evaluate and advance the state-of-the-art in protein function prediction. Here are some of the best approaches that have been used to solve the CAFA challenge:

\n

Machine learning-based methods: These methods use various machine learning algorithms such as Support Vector Machines (SVMs), Random Forests, and Deep Neural Networks (DNNs) to predict protein functions. These algorithms learn from previously annotated data and make predictions based on similarities between the target protein and annotated proteins.

\n

Network-based methods: These methods use protein-protein interaction networks to predict protein functions. They take advantage of the fact that functionally related proteins tend to interact with each other more frequently than with other proteins.

\n

Homology-based methods: These methods predict the function of a protein based on its sequence similarity to proteins with known functions. They assume that proteins with similar sequences tend to have similar functions.

\n

Hybrid methods: These methods combine multiple approaches, such as machine learning and network-based methods, to improve the accuracy of function predictions.

\n

Ensemble methods: These methods combine the results from multiple methods to produce a consensus prediction. This approach has been shown to improve the accuracy of function predictions.

\n

It's worth noting that the best approach for a specific problem in the CAFA challenge can depend on the specific features of the data being analyzed. Therefore, it's common to use a combination of multiple methods and to continuously refine them based on the results obtained in the challenge.

" }, { "id": 2226996, "authorName": "Marília Prata", "votes": 2, "postDate": "2023-04-19T12:56:19.483000", "content": "

Hi Alexander,

\n

I edited my comment since I read it too fast.

\n

After analysing critically the last paragraph of ChatGPT (\"depend on the specific features\". \"Use a combination of multiple methods\". \"Continuously refine them based on the results obtained\")

\n

Even with my poor knowledge, I thought ChatGPT didn't add anything concrete(on that last paragraph) for approaching the competition. The answer is abstract/vague and it could be applyed for answering any other challenge.

\n

Yesterday, when I checked ALS Challenge I realized that you are The Alexander(with the generic bird image) that had published so many remarkable Notebooks on that competition. And on Multimodal Single-Cell you had already changed your name and image.
\nGlad to know that since it's easier to recognize your brilliant, generous work.

\n

I'll be waiting to read your contributions on this competition.

" }, { "id": 2325222, "authorName": "compdad", "votes": 1, "postDate": "2023-07-01T07:27:32.157000", "content": "

Thanks for sharing. This is exactly what I was looking for!!

" }, { "id": 2325570, "authorName": "Marília Prata", "votes": 1, "postDate": "2023-07-01T13:02:50.577000", "content": "

Thank you for the kind words and support (mugichaa). Nice account name CompDad : )

" }, { "id": 2228129, "authorName": "waticson", "votes": 1, "postDate": "2023-04-20T09:37:11.687000", "content": "

Thank you for the informative list @mpwolke :)

" }, { "id": 2228264, "authorName": "Marília Prata", "votes": 1, "postDate": "2023-04-20T11:52:26.957000", "content": "

Thank you Waticson. For those that intend to learn Bioinformatics, it's a good start on Kaggle.

" }, { "id": 2227075, "authorName": "Marcel Ferreira", "votes": 1, "postDate": "2023-04-19T13:50:41.263000", "content": "

Thanks for sharing Maria.
\nVery cool to see Brazilian from bioinfo here on Kaggle!

\n

Nice to meet you!

" }, { "id": 2227144, "authorName": "Marília Prata", "votes": 1, "postDate": "2023-04-19T14:29:06.923000", "content": "

Oi Marcel,
\nobrigada pelo apoio. Estou apenas tentando aprender algo sobre programação no Kaggle.
\nBioinformatics eu sei menos ainda. Até para abrir os arquivos eu sofri, lendo dicas no StackOverflow.

\n

Porém, é um assunto extremamente interessante.

" }, { "id": 2227159, "authorName": "Marcel Ferreira", "votes": 1, "postDate": "2023-04-19T14:43:31.153000", "content": "

Que legal!

\n

Eu trabalho com pesquisas utilizando bioinfo! É realmente um campo sensacional!

\n

Estou a disposição para trocar informações se quiser!

\n

Abraço!

" }, { "id": 2226625, "authorName": "serangu", "votes": 1, "postDate": "2023-04-19T05:10:01.567000", "content": "

@mpwolke Very interesting list of helpful materials about the bioinformatic's theme.

" }, { "id": 2227005, "authorName": "Marília Prata", "votes": 0, "postDate": "2023-04-19T13:05:01.537000", "content": "

Indeed, that material is valuable for such an important subject as Bioinformatics.

" }, { "id": 2228938, "authorName": "yunsuxiaozi", "votes": 0, "postDate": "2023-04-21T00:35:35.397000", "content": "

Hello, may I ask if this competition requires a basic understanding of biology? Without knowledge of biology, is it not suitable to participate in this competition?

" } ], "index": { "id": "402597", "title": "Bioinformatics on Kaggle. Biopython.", "authorName": "", "commentCount": "12", "votes": "46", "postDate": "2023-04-19 01:14:13.113000" }, "competition": "cafa-5-protein-function-prediction" }