Update report according to Simon's Comments

abduskhazi / PL-Binding-Affinity-Prediction-using-ML

This repository is maintained for the documentation and coding of the MSc project @ Bioinformatics Lab Uni-Freiburg.

MIT License

1 stars 0 forks source link

Update report according to Simon's Comments #1

Open abduskhazi opened 3 years ago

abduskhazi commented 3 years ago

simon_comments.pdf Please check the comments in this pdf.

abduskhazi commented 3 years ago

Some comments regarding the report given

[x] fpocketp, fpocketnp and explicitp are 3 different types of pockets and not descriptors. The way I have written it creates confusion for the reader. Solution - mention that they are different types of pockets and write another section about the descriptors of ligands.
[x] Explain what SMILES is and elaborate a little bit more on SDF and mol2 so that the reader gets acquainted with the formats.
[x] Mention the R^2 score for each image and explain what the images represent and how to read them.
[x] Mention labels for figures i.e Figure 1, Figure 2 etc. This can be used to cross reference. for describing the figures.
[x] Create a new subsection "Reproducibility" Inside the TESTING STRATEGY section to explain how to reproduce the results exactly. Explain what execution ID represents.
[x] Write a short introduction to random forest regression before mentioning the hyper parameters used.

abduskhazi commented 3 years ago

Hi All, This ticket is now completed. More over, I have spent quite some time organising the report so that all the topics I have worked on are clearly explained. Please check the report.pdf Regards, Abdus Salam Khazi

simonbray commented 3 years ago

Thank you @abduskhazi, I'll read through tomorrow before our meeting.

abduskhazi commented 3 years ago

Tasks:

[x] Check the pdf attached with latest comments
[x] Check Papers available for fpocket, PDB, PDBBind, RDKit and include them in the references.
[x] Add latest findings in the report.

Major Comment from Simon - References section should contain more scientific literature and fewer online resources. report050821_Simon_comments.pdf

abduskhazi commented 3 years ago

Change PDB databank. --> PDB bind databank. The (7.4) error section to discussion. 7.5 could be put into the results section (possibly). References slide - 3 to 4 most important. No point in have all the references in the final slide. WE should have 3000 samples for test cases. (idea).

In the presentation after the problem definition give the entire workflow. (to give a bigger picture). Put this before feature feature extraction slide. Data preprocessing -> feature seleciton -> dealing with measurement resolution -> building model -> testing strategy -> results -> discussion.