Interpretation of compound fragments via attentive recursive tree

Simple item page
dc.contributorGraduate Program in Computational Science and Engineering.
dc.contributor.advisorÜlgen, Kutlu Ö.
dc.contributor.advisorÖzgür, Arzucan.
dc.contributor.authorÖzel, Nural.
dc.date.accessioned2025-04-14T12:09:52Z
dc.date.available2025-04-14T12:09:52Z
dc.date.issued2023
dc.description.abstractThe discovery of new drug-like chemicals with desired properties is a challenging and costly process in the pharmaceutical industry. To facilitate this process in the preclinical phase, many different neural network models have been proposed for different tasks (e.g., drug-target affinity prediction, molecular property prediction, targetspecific molecule generation). Despite producing successful results, they usually lack interpretability. To comprehend the significance of each fragment in the relevant compounds, we employed the Attention Recursive Tree (AR-Tree) model. Thanks to its task-specific attention mechanism, AR-Tree highlights the significant fragments of compounds by positioning them closer to the root of the tree structure. In this way, the identified significant fragments can be used to design new compounds with desired properties in future research. We experimented with five different classification and four different regression tasks of the MoleculeNet as benchmark tasks. The results of the experiments show that the proposed architecture succeeded in finding chemically meaningful fragments for the corresponding tasks.
dc.format.pagesxvii, 90 leaves
dc.identifier.otherGraduate Program in Computational Science and Engineering. TKL 2023 U68 PhD (Thes AD 2023 I81
dc.identifier.urihttps://digitalarchive.library.bogazici.edu.tr/handle/123456789/21502
dc.publisherThesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2023.
dc.subject.lcshChemicals.
dc.subject.lcshDeep learning (Machine learning)
dc.titleInterpretation of compound fragments via attentive recursive tree

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
b2795767.038435.001.PDF
Size:
5.65 MB
Format:
Adobe Portable Document Format
Download

Collections

M.S. Theses