English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Machine learning models have found numerous successful applications incomputational drug discovery. A large body of these models represents moleculesas sequences since molecular sequences are easily available, simple, andinformative. The sequence-based models often segment molecular sequences intopieces called chemical words (analogous to the words that make up sentences inhuman languages) and then apply advanced natural language processing techniquesfor tasks such as $\textit{de novo}$ drug design, property prediction, andbinding affinity prediction. However, the chemical characteristics andsignificance of these building blocks, chemical words, remain unexplored. Thisstudy aims to investigate the chemical vocabularies generated by popularsubword tokenization algorithms, namely Byte Pair Encoding (BPE), WordPiece,and Unigram, and identify key chemical words associated with protein-ligandbinding. To this end, we build a language-inspired pipeline that treats highaffinity ligands of protein targets as documents and selects key chemical wordsmaking up those ligands based on tf-idf weighting. Further, we conduct casestudies on a number of protein families to analyze the impact of key chemicalwords on binding. Through our analysis, we find that these key chemical wordsare specific to protein targets and correspond to known pharmacophores andfunctional groups. Our findings will help shed light on the chemistry capturedby the chemical words, and by machine learning models for drug discovery atlarge.

Exploring Data-Driven Chemical SMILES Tokenization Approaches to Identify Key Protein-Ligand Binding Moieties