Network analysis of
biological data
Gang Fang, PhD
New York University Shanghai Campus
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Guidelines and tips
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Insights from E. coli and Linux: Comparing Bio and Computer OS Networks, Slides of Biology
The concept of biological network analysis using the examples of E. coli transcriptional regulatory network and the Linux call graph. The author discusses the topology and evolution of these networks, their hierarchical organization, and the correlation between reuse and persistence. The document also highlights the differences between biological and computer operating systems networks in terms of modularity, node reuse, and robustness.
Typology: Slides
2021/2022
1 / 11
This page cannot be seen from the preview
Don't miss anything!
Related documents
Partial preview of the text
Download Insights from E. coli and Linux: Comparing Bio and Computer OS Networks and more Slides Biology in PDF only on Docsity!
Network analysis of
biological data
Gang Fang, PhD
New York University Shanghai Campus
How much data is there in biology?
How big is the cell?
How thick is the cell wall?
How long is its flagellum?
How long does the cell live?
How much “nutrition” does the cell need?
How long is the genome?
How many genes are there in the genome?
How many mRNAs inside the cell?
How many proteins inside the cell?
How many metabolites inside the cell?
How long does one mRNA “stay” in the cell?
How much “energy” does it take to synthesize protein?
Is the data constant for all time, under all circumstances?
What's the best way to describe biology?
What is biological network
- Node
Gene, protein, metabolite, any “biological object”
- Edge
Regulation, protein-protein interaction, any kind of “similarity” or “dissimilarity” etc.
- “Weights” or features
Conservation of gene, expression value, half time, any measurable or categorical variable.
- Network topology
Clusters, modularity, node centrality, shortest path etc.
- Network dynamics
Comparison of networks: time series and environmental changes
- Network rewiring and permutation
Test your theory!
Hierarchical organization:
pyramidal versus top-heavy
E. coli transcriptional regulatory network the Linux call graph
master regulator
workhorse
middle manager
Persistent genes^ Persistent functions
Genes subject to strong natural selective pressure Software engineers’ favorite functions
Organization of Modules:
independent versus overlap
7
E. Coli TRN
Linux call graph Average overlap 4.3% 80.7% Maximum node reuse 15.6%^ 87.5% Average node reuse
Modules are labeled by master regulators: TFs, high-level starting functions
TRN:
modules overlap little, components are less generic
M 2 ∩ M 3 M 2 ∪ M 3
= 2 11
Overlap(M2,M3)=
reuse=2/3 reuse=1/
M M2 (^) M
Call graph: modules overlap, Functions are highly reused (generic): “printk”
We observe opposite correlation behaviors in the two systems: Reuse and persistence are negatively correlated in the E. coli regulatory network but positively correlated in the Linux call graph.
[Spearman correlation r=−0.074 (P < 0.01) and r=0.10 (P < 10 −4), respectively]
Can we use network analysis to
identify protein “living fossils”?