5. Allow organizations to
support multiple graphs
in their Neo4j footprint
5
Allow organizations to
store more sensitive
data in their graphs
Neo4j 4.1
Allow organizations
to scale their Neo4j
infrastructure
7. Competitive Intelligence
Compounds
Research
Compounds
Data Sources:
● Worksheets
● PubMed
● CT.gov
● Google
Patents
● FDA
● Academic
Papers
● Grant
Applications
Research
Compounds
Research
Use
r
Compounds
Workb
ooksClinical
Trials
Patents
Research
Workb
ooks
Workb
ooks
Workb
ooks
FDA
14. Who can access the medical information for
patients during a clinical trial?
How do I set up permissions so the right
individual can see the correct data?
● Read vs. Write Access in Graphs
Do I want my administrators to have access
to data?
14
Patient Information Security and
Compliance
23. Graph Data Science is a
science-driven approach to gain
knowledge from the relationships
and structures in data, typically to
power predictions.
23
What is Graph data science?
Data scientists use
relationships to answer
questions.
24. Query (Cypher)
Real-time, local decisioning
and pattern matching
Graph Algorithms
Global analysis
and iterations
You know what you’re
looking for and making a
decision
You’re learning the overall structure
of a network, updating data, and
predicting
Local
Patterns
Global
Computation
25. Robust Graph Algorithms
• Run on a loaded graph to compute metrics about the
topology and connectivity
• Highly parallelized and scale to 10’s of billions of nodes
25
The Neo4j GDS Library
Mutable In-Memory
Workspace
Computational Graph
Native Graph Store
Efficient & Flexible Analytics Workspace
• Automatically reshapes transactional graphs into
an in-memory analytics graph
• Optimized for analytics with global traversals
and aggregation
• Create workflows and layer algorithms
26. Evolution of Graph Data Science
Decision
Support
Graph Based
Predictions
Graph Native
Learning
26
Graph Feature
Engineering
Graph
Embeddings
Graph
Networks
Knowledge
Graphs
Graph
Analytics
27. Evolution of Graph Data Science
Graph Feature
Engineering
Graph
Embeddings
Graph
Networks
27
Graph
AnalyticsKnowledge
Graphs
Graph search
and queries
Support domain
experts
28. Deceptively Simple Queries
What completes the connections from
genes to diseases to targets?
What genes can be reached +4 hops
out from a known drug target?
What mechanisms in common are
there between two drugs?
Which biological pathways are
regulated by these genes?
Knowledge Graph Queries in Drug Discovery
Connecting concepts like genes, chemicals, and diseases lets you link and mine
interconnected concepts like genes, diseases, and chemicals.
28
29. Evolution of Graph Data Science
Graph Feature
Engineering
Graph
Embeddings
Graph
Networks
29
Knowledge
Graphs
Graph
Analytics
Graph queries &
algorithms for
offline analysis
Understanding
Structures
30. Graph Algorithms for Drug Discovery
PageRank & Betweenness to
identify essential regulatory
genes or drug targets
Louvain to identify protein
regulatory networks
Shortest path to link drug
targets to possible outcomes
or side effects
Node Similarity to find
structurally similar chemicals
Link Prediction to estimate
likelihood of interactions
Identity drug mechanisms and new targets based on network structure
31. Evolution of Graph Data Science
Graph
Embeddings
Graph
Networks
31
Knowledge
Graphs
Graph
Analytics
Graph Feature
Engineering
Graph algorithms
& queries for
machine learning
Improve Prediction
Accuracy
32. Graph Feature Engineering in Drug Discovery
Feature Engineering is how we combine and process the data
to create new, more meaningful features. Using graphs we
can base ML on genes with similar network topology.
32 Gene and Target Data
33. Evolution of Graph Data Science
Decision
Support
Graph Based
Predictions
Graph Native
Learning
33
Graph Feature
Engineering
Graph
Embeddings
Graph
Networks
Knowledge
Graphs
Graph
Analytics
FUTUREEarly Adopters
35. The Data
35
Synthetic electronic medical records
from
- Open source, synthetic patient
generator developed by the MITRE
corporation
- Flexibly generates medical history
of synthetic patients
- Intended to be realistic but not real
- data distribution and correlations
mimic real EHRs
code & data are available at github.com/AliciaFrame/GDS_patient_journey_demo
36. Thank You
&
Questions
36
- O’Reilly Book on Graph Algorithms
neo4j.com/graph-algorithms-book/
- GDS website, whitepapers, links
neo4j.com/use-cases/graph-data-s
cience-artificial-intelligence/
- Patient Journey Demo:
github.com/AliciaFrame/GDS_Patie
nt_Journey
- GDS Sandbox sandbox.neo4j.com/
?usecase=graph-data-science
Resources
Pat Wall
patrick.wall@Neo4j.com
Alicia Frame
Alicia.Frame@Neo4j.com
@AliciaFrame1