The Diversity and Integration of Biological Network Motifs Seminars in Bioinformatics Martin Akerman 31/03/08

Biological Networks Questions: Which are the most common motifs among biological networks? How do these motifs interrelate?

Biological Network motifs BiFan Diamond Autoregulation (AR) Feed Forward Loops (FFL) Regulating and Regulated Feedback Loops (RFL) Single Input Model (SIM) Dense Overlapping Regulon (DOR) Cascade

The SIMs are common in sensory transcription networks: Genes from a same Pathway (Arginine synthesis). Genes responding to stress (DNA repair). Genes that assemble a same biological machine (ribosomal genes). Single Input Model (SIM)

The SIMs can generate temporal programs of expression: Single Input Model (SIM) Last-In First-Out (LIFO) Program

LIFO Program in Arginine Biosynthesis

First-In First-Out (FIFO) Program K xz1 >K xz2 >K xz3 K’ xz1 <K’ xz2 <K’ xz3 Time K xz1 K xz2 K xz3 K xz2 K xz1 [X] [Y] [Z 2 ] [Z 3 ] [Z 1 ]

FIFO program in Flagella Biosynthesis

FIFO program is governed by a FFL

Multi-input FFL in Neuronal Networks FLPASH AVD AVA Nose Touch Noxious Chemicals Nose Touch Backward movement

Multi-input FFL in Neuronal Networks The change in voltage of Y The change in voltage of Z Y X1 X2 Z

Interlocking Feed forward loops Bacillus Subtilis sporuation process

Dense Overlapping Regulon (DOR)

How do Network Motifs Integrate? The E.coli Transcription Network (partial) A single DOR Layer FFLs and SIMs are integrated within DORs A Master Regulators Layer (lots of Auto-Reg.) Where are the X  Y  Z ?

Signal Transduction Cascades

Popular Motifs in Signal Transduction Cascades X1X2 Y1Y2 Generalization of DOR BiFan Y2 Z Y1 Diamond X1X2 Y1Y2 Z Y1Y2 Z1Z2 X Y1Y2 Z1Z2 X1X2 Multi-layer Perceptrons (multi-DORs) X

X1 P Y1 P Z1 P X2 P Y2 P Z2 P Multi Layer Perceptorns in Signal Transduction Cascades

Dynamics of Signal Transduction Cascades At Steady State, Activation Threshold Y X1 X2 X1 X2 0.5

Dynamics of Signal Transduction Cascades Y X1 X2 X1 X2 0.5 “AND” gate Y X1 X2 X1 X2 0.5 “OR” gate

Dynamics of Signal Transduction Cascades X1X1 X2X2 Y1Y1 Y2Y2 Z X1X1 X2X2 Y1Y1 Y2Y2 Z “AND” gate “OR” gate Y1Y1 Y2Y2 Z Y1Y1 Y2Y2 Z

Dynamics of Signal Transduction Cascades X1X1 X2X2 Y1Y1 Y2Y2 Z X1X1 X2X2 Y1Y1 Y2Y2 Z Y1Y1 Y2Y2 Z Y2Y2 Z Y1Y1 Z

Multi-layer perceptrons can show:  Discrimination : the ability to accurately recognize certain stimuli patterns.  Generalization : the ability to fill the gaps in partial stimuli patterns.  Graceful degradation : damage to elements of the perceptron or it connections does not bring the network to crashing halt Dynamics of Signal Transduction Cascades

Feed Back Loops XY Z (Fast) Protein-Protein Interactions (Slow) Transcriptional Interactions Z transcriptionally activates X and Y X forms a complex with Y. X phosphorylates Y. Y X X transcriptionally activates X. Y inhibits X. PowerHeater Thermostat Temperature -

Feed Back Loops Produce Oscillation Mutation of the Drosophila CWO gene Cdc20 oscillator controls Cell Cycle

Developmental Transcription Netwroks The TF expression profile in a developing Drosophila embryo

Developmental Transcription Netwroks X Y X Y Both X AND Y are ON at the same time. Genes regulated by X and Y belong to the same tissue (or strip). X OR Y is ON at a given time. Genes regulated by X and Y belong to different tissues (strips). Two-node Feedback Loops

Developmental Transcription Netwroks XY Z XY Z XY Z XY Z Regulating Feedback Loops Regulated Feedback Loops Double Positive LoopsDouble Negative Loops

Developmental Transcription Netwroks Regulated Feedback Loops as a Memory Element

Developmental Transcription Netwroks Cascades XYZXYZ

Summary Network motifs can function in  several biological processes (sensory systems, development).  different time scales (milliseconds, cell generations). Network motifs can produce temporal programs (LIFO, FIFO, oscillation). Motifs within a network may be arranged in organized structures (perceptrons, interlocking FFL). Different kinds of network may interact to generate regulation