5 min for questions Alliance for Cellular Signaling Pasadena, CA May 19, 2003Monday Morning Single and Double Ligand Screens 9:00 Rama Ranganathan Single.

5 min for questions

Alliance for Cellular SignalingPasadena, CAMay 19, 2003 Monday Morning

Single and Double Ligand Screens9:00 Rama Ranganathan

Single ligand screen: What did we do; what did we learn? Signaling experiments

9:25 Mel Simon Single ligand screen: What did we do; what did we learn? Microarray experiments

9:55 Break10:25 Paul Sternweis

Double ligand screen: experimental design and initial patterns of interaction

10:55 Rama RanganathanAnalysis and results of double ligand screen data

11:20 Discussion 11:50 Wrap-up

Ligand ScreensAfCS Goal: Quantitative understanding of the

signaling network in a mammalian cell

1. Independent of cell type2. Focused on pathway structure and interactions3. Complex system: numerically, spatially and organizationally 4. Includes processes in multiple time domains5. Includes both mass-action and stochastic processes

The ligand screen is our first real experiment

Ligand ScreensAfCS Goal: Quantitative understanding of the

signaling network in a mammalian cell

1. Independent of cell type2. Focused on pathway interactions3. Complex system: numerically, spatially and organizationally 4. Includes processes in multiple time domains5. Includes both mass-action and stochastic processes6. Not amenable to simply exhaustive data gathering7. Requires black box (top down) models

to direct choice of cellular signaling probes 8. Requires black box (behavioral) data for experimental design and,

later, to evaluate bottom-up models

Experiments will depend on general understanding of the nature and exent of the complexity of our cells


Ligand ScreensWhat behaviors lie within the black box?

What are its rules?

Challenge: acquire, organize and interpret a data set that will allow us to

choose the cellular signaling probes that willbe most informative

assemble and evaluate input/output data constrain models of how the signaling

pathways work in cells according to overall cellular behaviors


single-ligand screen is...

Single Ligand Screen

How many inputs are there?

How many patterns of response are there?

What are the structures of these patterns?

How do they differ?

How are these patterns organized?


How many inputs are there ?



How do they differ?


Are there rules for network organization?

What are the rules?

How many regulatory ligands does a cell respond to ?How many do we have to look at ? What is the most efficient way to look ? Depends on # of

signaling patterns





How do they differ?



What are the rules?

What ligands are redundant?What ligands give overlapping outputs?How many ligands with “identical” outputs should

we consider?How do we identify “identical” experimentally?

Can response patterns be productively clusterd ?





How do they differ?



What are the rules?

Can we identify conserved patterns of output patternseven if individual outputs differ?

How do we look for them experimentally?Do different response patterns draw from a listable

toolbox of signal outputs?Do different response patterns reflect

branched pathways, mulitple receptors, ...?





How do they differ?



What are the rules?

Distinctive groupings of individual responses? Relative dynamics and time courses?





How do they differ?



What are the rules?

How does a cell assemble its signaling network?Buzzwords: module, heirarchy, parallelHow do you rigorously identify such patterns?

Double Ligand Screen

RA+B = RA + RB

?

Conceptual violations of signal additivity

0+0=1 1+1=0 1+0=0 1+1=17 1+0=17

How often do two signaling pathways interact

to produce a response other than the sum of

the two individual responses


How numerous (dense) are ligand - ligand interactions?

Which ligands interact?

Are interactions determined primarily by individual outputs?

How do you identify such interactions conceptually?

How do you identify them experimentally?

What are the mechanisms of these interactions?

Are the interactions regulated? (Triple-ligand screen...)

Are there rules for network organization? What are they?

RA+B = RA + RB

?



RA+B = RA + RB

?

How many ligand pairs must we consider?How complex a network should we plan on studying?How do we prepare our models to direct our experimental design? to map such a system?

Does a cell process information more through complex interactions than with multiple signals ?





RA+B = RA + RB

?

Will receptors with similar individual outputs demonstratesimilar spectra of double-ligand interactions?

How does this result impact on our experiments?






RA+B = RA + RB

?

What do “additive” and “non-additive” mean in systems that saturate?

Cooperativity, topping out, competition, inhibition,cross-talk, weirdness ...






How do you identify them experimentally?

What are the mechanisms of these interactions?

Are the interactions regulated? (Triple-ligand screen...)

Are there rules for network organization? What are they?

RA+B = RA + RB

?

• established strategies for the single ligand screen.• set up some of the wide array of necessary assays.• executed a single-ligand screen in B cells and WEHI.• begun to analyze single-ligand data.

provocative enough that it won’t be “done” soon• established strategies for the double-ligand screen.

non-trivial and still under developmenttime and concentration domains different kinds of interactionschoice of measured outputschoices influenced by types of assays

• run partial double-ligand screen on B cells.

How do we best and most efficiently execute and analyze the single- and double-ligand screen on a new cell type?

So far, we have...

Title

Single and Double Ligand Screens

What’s next?

•Evaluate applicability of microarrays to double-ligand screen

Lots of data; not quite quantitative; expensive

Replace with a few well-chosen RT-PCR’s?

•Bring multiplex lipid analysis on-line

Extend to inositol lipids

•Adopt standards for unified data normalization

•Develop and adopt formulas to choose ligand concentrations

•Automate data normalization and integration

•Add new assays as feasible

Criteria for New AfCS Assays

• Highly multiplexed or complex (integrative) readouts• Quantitative data• Large dynamic range• Reproducible• High throughput • Responsive to many inputs ( or )• Continuous time-base readout 15 min• Assorted depths into pathways• Single cell measurements, but lots of them• Data obtainable in < 12 months; established assays

Title

Single and Double Ligand Screens

What’s next?

•Evaluate applicability of microarrays to double-ligand screen

Lots of data; not quite quantitative; expensive

Replace with a few well-chosen RT-PCR’s?

•Bring multiplex lipid analysis on-line

Extend to inositol lipids

•Adopt standards for unified data normalization

•Develop and adopt formulas to choose ligand concentrations

•Automate data normalization and integration

•Add new assays as feasible

Initiate the screens on our new cell type

Title

Alliance for Cellular SignalingPasadena, CAMay 19, 2003

Single and Double Ligand Screens9:00 Rama Ranganathan

Single ligand screen: What did we do; what did we learn? Signaling experiments

9:25 Mel Simon Single ligand screen: What did we do; what did we learn? Microarray experiments

9:55 Break10:25 Paul Sternweis

Double ligand screen: experimental design and initial patterns of interaction

10:55 Rama RanganathanAnalysis and results of double ligand screen data

11:20 Discussion 11:50 Wrap-up

5 min for questions Alliance for Cellular Signaling Pasadena, CA May 19, 2003Monday Morning Single and Double Ligand Screens 9:00 Rama Ranganathan Single.

Documents

patterns of response

response patterns

double ligand screens

mel simon single ligand

ligand screens afcs

real experiment slide

morning single

wrapup slide