Bioactivity Data : Binding database and DrugBank

Bioinformatics, topics covered are bioactivity data, BindingDB, and drug bank databases, how to use the database, insights, and rules

1. Bioactivity data:
BindingDB &
DrugBank

2. LIST OF
CONTENTS
BINDINGDB VS. DRUGBANK:
A COMPARISON
06
01
THE DRUG DISCOVERY
PROBLEM
02
WHAT IS BIOACTIVITY DATA?
03
THE DATA EXPLOSION →
WHY DATABASES?
04
A LOOK INSIDE BINDINGDB
05
INTRODUCING DRUGBANK
07
CONCLUSION & THE FUTURE

3. WHAT IS
BIOACTIVITY?

4. Cost : Expensive ??????
THE DRUG DISCOVERY
CHALLENGE
Timeline : 10-15 Years ⌛
Sucess Rate: < 10% ,risky ❌
The Goal : Find the right key for the
lock ??????

5. Bioactivity data
The measurable output of drug-target interactions.
Kd (Dissociation Constant):
Strength of the complex formation
Lower value = Tighter bind ??????
Binding Affinity (Strength)
Ki (Inhibition Constant):
Strength of inhibition,
Lower value = More potent inhibitor ??????
The essential fuel for AI-driven drug discovery & design
(How tight is the handshake?) ??????
Binding Affinity (Strength)
(How tight is the handshake?) ??????K(Dissociation Constant):
Strength of the complex formation
Lower value = Tighter bind ??????
d K (Inhibition Constant): i
Strength of inhibition,
Lower value = More potent inhibitor ??????
Potency (Dose)
(How much drug is needed???????)
IC₅₀ (Half-Maximal Inhibitory Concentration):
Amount of drug needed to block 50% of activity.
??????
Lower value = More potent inhibitor
EC₅₀(Half-Maximal Effective Concentration):
Amount of drug needed to cause 50% of effect. ↗️
Lower value = More potent activator

6. THE NEED FOR CENTRALIZED
KNOWLEDGE The Problem: Data Overload
Millions of data points buried in
patents & papers ??????
Manually tracking it is inefficient
and impossible ??????‍?????? The Solution: Structured Databases
Centralized, curated, and computable
knowledge ??????

7. The Specialist: Public database for
quantitative binding data ??????
Introducing
BindingDB
How tight is the handshake?" ??????
how strongly molecules bind to
proteins
Its Scale: +3.1M measured binding
interactions
Can be accessed via:
https://www.bindingdb.org/rwd/bin
d/index.jsp.

8. Idea first discussed
at a NIST + Rutgers
University
workshop
1997
Project formally
initiated at Rutgers
University (Prof.
Michael Gilson’s
lab).
2000
First public release
of BindingDB
(focused on
enzyme–inhibitor
data).
2004
Expanded with
thousands of
ligand–target
interactions.
2007
Major growth:
added patent data,
integrated with
PubChem, UniProt,
ChEMBL, PDB.
2010–2015
New version
released with
broader search
tools + REST API
access
2016
TIMELINE OF BINDINGDB??????‍??????
?????? ?????? ??????
??????
??????

9. TIMELINE OF BINDINGDB
Special COVID-
19 datasets
added for
global
researchers.
2020
Surpassed 2.5
million binding
measurements
milestone.
2023
Contains ~3.1M binding
measurements, 1.34M
compounds, 9,600
targets; updated monthly.
2025 (NOW)
?????? ??????

11. Commitment to the FAIR Principles
Findable (F):
Each entry has a unique identifier → makes searching simple
Accessible (A):
Available to all users; long-term preservation ensured by UC San Diego
Library archiving
Interoperable (I):
Links with PubChem, UniProt, PDB using standardized formats → data works
across platforms
Reusable (R):
Entries fully annotated with experimental details; shared under CC BY 4.0
license → reuse allowed for academic & commercial use with credit

12. INTRODUCING
DRUGBANK
The Encyclopedia: Comprehensive
knowledge hub for drugs & drug targets ??????
Core Question: "What does this drug do
in the body?" ??????️‍??????️
Its Scale: ~16,000 drug entries with
~100 data fields each

13. The DrugBank :
A 360° View

15. BindingDB DrugBank
Primary Focus Quantitative Binding Affinities
Comprehensive Drug
Knowledge
Core Data
Kd, Ki, IC₅₀ (Strength & Potency) Chemistry, Pharmacology,
Clinical Info
Key Question How tight is the handshake?
What does it do in the
body?
Best For
Lead Optimization, AI/ML Training
Drug Repurposing, Clinical
Decision Support
BindingDB vs DrugBank

16. CONCLUSION
Bioactivity data is the cornerstone of
modern drug discovery
BindingDB (the specialist) + DrugBank
(the generalist) = A powerful,
complementary workflow
The future lies in leveraging AI to
integrate these data sources, enabling
faster and smarter drug development

17. Any Questions

18. Presented by:
Sree