Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Drug Discovery: Overview01:26

Drug Discovery: Overview

8.5K
Drug discovery is a multifaceted process involving extensive screening, testing, and optimization of lead compounds to identify potential new drugs for therapeutic use. It combines several approaches, including screening large numbers of natural products, chemical modification of known active molecules, identification of new drug targets, and rational design based on biological mechanisms and drug-receptor structure. These approaches are carried out in both academic research laboratories and...
8.5K
Structure-Activity Relationships and Drug Design01:28

Structure-Activity Relationships and Drug Design

988
Drug design is a dynamic field that involves discovering and developing new medications based on specific biological targets. This process heavily relies on structure-activity relationships (SAR) and quantitative structure-activity relationships (QSAR) to guide the design and optimization of efficient drugs.
SAR studies the intricate relationship between a drug's chemical structure and biological activity. It focuses on understanding how modifications to a drug's structure can influence...
988
Targets for Drug Action: Overview01:26

Targets for Drug Action: Overview

7.2K
Drugs target macromolecules to modify ongoing cellular processes. Primary drug targets include receptors, ion channels, transporters, and enzymes.
Receptors are either membrane-spanning or intracellular proteins, which upon binding a ligand, get activated and transmit the signal downstream to elicit a response. Drugs bind receptors, either mimicking the action of endogenous ligands or blocking the receptor activity to bring about a modified response. Nearly 35% of approved drugs target the G...
7.2K
Prodrugs01:30

Prodrugs

2.7K
Prodrugs are a class of pharmaceutical compounds that undergo a biotransformation process within the body to be converted into a pharmacologically active drug. Prodrugs are designed to improve the therapeutic properties of the parent drug, such as enhancing bioavailability, increasing stability, or reducing toxicity. The concept of prodrugs revolves around modifying the chemical structure of the original drug to make it more effective or convenient for administration.
Prodrugs help overcome...
2.7K
Ligand Binding Sites02:40

Ligand Binding Sites

13.1K
Proteins are dynamic macromolecules that carry out a wide variety of essential processes; however, the activities of most proteins depend on their interactions with other molecules or ions, known as ligands.
Protein-ligand interactions are quite specific; even though numerous potential ligands surround a cellular protein at any given time, only a particular ligand can bind to that protein. Moreover, a ligand binds only to a dedicated area on the surface of the protein, known as the...
13.1K
Drug Biotransformation: Overview01:16

Drug Biotransformation: Overview

2.6K
Pharmaceutical substances known as xenobiotics are predominantly lipophilic and nonionized. This enables them to permeate lipid bilayers, such as cell membranes, and interact with intracellular target receptors. Lipophilic drugs have an advantage in crossing biological barriers and reaching their intended sites of action. However, lipophilic drugs often have a restricted capacity for renal expulsion or elimination from the body. When these drugs enter the kidneys and undergo glomerular...
2.6K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

PUFFIN: protein unit discovery with functional supervision.

Bioinformatics (Oxford, England)·2026
Same author

Novel 2-indolinone benzylpiperidine- thiosemicarbazone hybrids as cholinesterase inhibitors: Design, synthesis and biological evaluation.

Bioorganic chemistry·2026
Same author

ConGen: Targeted Molecule Generation Through Contrastive Learning and Latent Optimization.

Molecular informatics·2026
Same author

VO: The Vaccine Ontology.

Scientific data·2026
Same author

Nested Named Entity Recognition using Multilayer BERT-based Model: Notebook for the BioASQ Lab at CLEF 2024.

CEUR workshop proceedings.·2026
Same author

Harnessing Phytochemicals for Brain Cancer Therapy: Insights Into Medulloblastoma and Glioblastoma Treatment.

Phytochemical analysis : PCA·2026

Related Experiment Video

Updated: Aug 28, 2025

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

541

Exploiting pretrained biochemical language models for targeted drug design.

Gökçe Uludoğan1, Elif Ozkirimli2, Kutlu O Ulgen3

  • 1Department of Computer Engineering, Boğaziçi University, İstanbul 34342, Turkey.

Bioinformatics (Oxford, England)
|September 20, 2022
PubMed
Summary

Warm-starting deep generative models with pretrained biochemical language models improves targeted molecule generation for drug design. The one-stage strategy showed better generalization in docking evaluations compared to the two-stage approach.

More Related Videos

Biosensor-based High Throughput Biopanning and Bioinformatics Analysis Strategy for the Global Validation of Drug-protein Interactions
08:31

Biosensor-based High Throughput Biopanning and Bioinformatics Analysis Strategy for the Global Validation of Drug-protein Interactions

Published on: December 1, 2020

5.1K
Author Spotlight: Streamlining Protein Target Prediction and Validation via Molecular Docking and CETSA
10:21

Author Spotlight: Streamlining Protein Target Prediction and Validation via Molecular Docking and CETSA

Published on: February 23, 2024

2.8K

Related Experiment Videos

Last Updated: Aug 28, 2025

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

541
Biosensor-based High Throughput Biopanning and Bioinformatics Analysis Strategy for the Global Validation of Drug-protein Interactions
08:31

Biosensor-based High Throughput Biopanning and Bioinformatics Analysis Strategy for the Global Validation of Drug-protein Interactions

Published on: December 1, 2020

5.1K
Author Spotlight: Streamlining Protein Target Prediction and Validation via Molecular Docking and CETSA
10:21

Author Spotlight: Streamlining Protein Target Prediction and Validation via Molecular Docking and CETSA

Published on: February 23, 2024

2.8K

Area of Science:

  • Computational chemistry and drug discovery
  • Artificial intelligence in pharmaceuticals
  • Bioinformatics and computational biology

Background:

  • Drug discovery relies on novel compounds targeting specific proteins.
  • Deep generative models show promise for targeted molecular design.
  • Limited protein-ligand pairs hinder current target-specific molecule generation models.

Purpose of the Study:

  • To leverage pretrained biochemical language models for targeted molecule generation.
  • To investigate two warm-start strategies: one-stage and two-stage.
  • To compare beam search and sampling decoding strategies for compound generation.

Main Methods:

  • Exploiting pretrained biochemical language models to initialize targeted molecule generation models.
  • Implementing a one-stage warm-start strategy (direct training).
  • Implementing a two-stage warm-start strategy (pre-finetuning then target-specific training).
  • Comparing beam search and sampling for compound generation.

Main Results:

  • Warm-started models outperform baseline models trained from scratch.
  • Both one-stage and two-stage strategies yield comparable results on benchmark metrics.
  • The one-stage strategy demonstrates superior generalization in docking evaluations for novel proteins.
  • Beam search is more effective than sampling for compound quality assessment.

Conclusions:

  • Warm-starting significantly enhances targeted molecule generation models.
  • The one-stage warm-start strategy offers better generalization for novel targets.
  • Beam search is the preferred decoding method for generating high-quality compounds.