Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences The power of interactivity December 10, 2014 Warren B. Powell Kris Reyes Si Chen Princeton.

Slides:



Advertisements
Similar presentations
Quality Series (Sample Slides)
Advertisements

Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences Richer belief models December 10, 2014 Warren B. Powell Kris Reyes Si Chen Princeton University.
EGR 334 Thermodynamics Chapter 6: Sections 1-5
Estimation and Reporting of Heterogeneity of Treatment Effects in Observational Comparative Effectiveness Research Prepared for: Agency for Healthcare.
PH403: Results Section Janet Tate Acknowledgements
RANSAC experimentation Slides by Marc van Kreveld 1.
Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences A case application – Growing carbon nanotubes December 10, 2014 Warren B. Powell Kris Reyes.
Lesson 5 Histograms and Box Plots. Histograms A bar graph that is used to display the frequency of data divided into equal intervals. The bars must be.
An Optimal Learning Approach to Finding an Outbreak of a Disease Warren Scott Warren Powell
Hierarchical Reinforcement Learning Ersin Basaran 19/03/2005.
© 2009 Warren B. Powell 1. Optimal Learning for Homeland Security CCICADA Workshop Morgan State, Baltimore, Md. March 7, 2010 Warren Powell With research.
Two Variable Analysis Joshua, Alvin, Nicholas, Abigail & Kendall.
Improving Stable Processes Professor Tom Kuczek Purdue University
Section 5: Graphs in Science
Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences Working with nonlinear belief models December 10, 2014 Warren B. Powell Kris Reyes Si Chen.
Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences Building a belief model December 10, 2014 Warren B. Powell Kris Reyes Si Chen Princeton University.
Topic A Factors Affecting Rates of Reaction
An Adaptive Modeling for Robust Prognostics on a Reconfigurable Platform Behrad Bagheri Linxia Liao.
Computational Stochastic Optimization: Bridging communities October 25, 2012 Warren Powell CASTLE Laboratory Princeton University
Observations & Measurements. Observations Information gained through our five senses Can be qualitative or quantitative.
SY PVAAS Scatter Plots State to IU Region to School District Grades 4-8, 11 Math & Reading PVAAS Statewide Team for PDE Contact your IU PVAAS contact.
Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences The knowledge gradient December 10, 2014 Warren B. Powell Kris Reyes Si Chen Princeton University.
Unit 1 The Basics of Biology. Goals of All Science Investigate and Understand the natural world Explain what happens in the natural world Predict what.
Chapter 1 Science Skills. 1.1 What is Science?  Science is a system of knowledge and the methods you use to find that knowledge  The goal of science.
Psy B07 Chapter 4Slide 1 SAMPLING DISTRIBUTIONS AND HYPOTHESIS TESTING.
Copyright © 2003 Pearson Education, Inc. Slide 11-0 Ch 11 Learning Goals 1.Operating, financial, and total leverage (causes & measures). 2.Business risk,
DOX 6E Montgomery1 Unreplicated 2 k Factorial Designs These are 2 k factorial designs with one observation at each corner of the “cube” An unreplicated.
Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences Overview December 10, 2014 Warren B. Powell Kris Reyes Si Chen Princeton University
© 2008 Pearson Addison-Wesley. All rights reserved Chapter 5 Statistical Reasoning.
Improvement Guide Workshops Chapters 1-14 Suggested Workshops for participants who are NOT working on a formal project as part of the workshop.
Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences Forming the decision set December 10, 2014 Warren B. Powell Kris Reyes Si Chen Princeton.
1 1 Slide Simulation Professor Ahmadi. 2 2 Slide Simulation Chapter Outline n Computer Simulation n Simulation Modeling n Random Variables and Pseudo-Random.
Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences Searching a two-dimensional surface December 10, 2014 Warren B. Powell Kris Reyes Si Chen.
Measurement. Why measure? Scientists use a standard method to collect data as well as use mathematics to analyze measurements. We must measure things.
Sequential Off-line Learning with Knowledge Gradients Peter Frazier Warren Powell Savas Dayanik Department of Operations Research and Financial Engineering.
Generation of anomalously energetic suprathermal electrons by an electron beam interacting with a nonuniform plasma Dmytro Sydorenko University of Alberta,
CHAPTER 2 STUDY GUIDE. WHAT NUMBER IS THE METRIC SYSTEM BASED ON? The number 10.
Chapter 9 Introduction to the t Statistic
Modeling promoter search by E.coli RNA polymerase : One-dimensional diffusion in a sequence-dependent energy landscape Journal of Theoretical Biology 2009.
Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences The value of information December 10, 2014 Warren B. Powell Kris Reyes Si Chen Princeton.
Plot Diagram.
BSHS 442 Week 2 DQ 1 Check this A+ tutorial guideline at 442/BSHS-442-Week-2-DQ-1 Describe the identified conflict.
ABS 417 Week 3 DQ 2 Social Science Research Check this A+ tutorial guideline at 3-dq-2-social-science-research.
Biology-Unit I-Part A The Scientific Method.
Step 1: Specify a null hypothesis
Optimal Learning in the Laboratory Sciences
Binding Thermodynamics of Ferredoxin:NADP+ Reductase: Two Different Protein Substrates and One Energetics  Marta Martínez-Júlvez, Milagros Medina, Adrián.
Israel maritime college
Statistics (0.0) IB Diploma Biology
Introduction The fit of a linear function to a set of data can be assessed by analyzing residuals. A residual is the vertical distance between an observed.
Department of Chemistry Princeton Univesity
RNA-Directed DNA Methylation: Getting a Grip on Mechanism
Let’s Organize the Data!
Gene quantification using real-time quantitative PCR
Data Analysis – Charts & Graphs
2. Stratified Random Sampling.
Unit 3 Science Investigation Skills
Graphing in Science.
Volume 59, Issue 1, Pages (July 2015)
Vocab Week 2 Mr. Addeo.
Volume 22, Issue 9, Pages (February 2018)
Volume 101, Issue 1, Pages (July 2011)
Introduction The fit of a linear function to a set of data can be assessed by analyzing residuals. A residual is the vertical distance between an observed.
Functional Role of Ribosomal Signatures
Optimal Learning in the Laboratory Sciences
Mapping the Position of Translational Elongation Factor EF-G in the Ribosome by Directed Hydroxyl Radical Probing  Kevin S Wilson, Harry F Noller  Cell 
Biophysical Characterization of Styryl Dye-Membrane Interactions
Chapter 5 Lecture Outline See PowerPoint Image Slides
Scatter Graphs.
Principles of Quantitative PCR
Presentation transcript:

Slide 1 Tutorial: Optimal Learning in the Laboratory Sciences The power of interactivity December 10, 2014 Warren B. Powell Kris Reyes Si Chen Princeton University Slide 1

Lecture Outline 2  The power of interactivity

Interactivity Guiding the experimental process requires balancing a number of objectives  We would like to run experiments with the highest chance of success.  We also want to learn the most to guide future experiments.  But we also have to consider the cost and complexity of an experiment. 3

Interactivity Prior estimate Value of information Temperature Pressure Highest value of information Temperature Pressure The experiment that produces the highest value of information might require using the highest temperatures and pressures. The scientist has to balance the performance of an experiment, the value of information, and the complexity of the experiment.

Interactivity The accessibility of a region of an RNA molecule mediates interaction with other molecules. Depends on how you define interactions, resulting in several methods for accessing accessibility. Mechanistic Energetic Mechanistically accessible region Well- protected region = inaccessible Designing probles for an RNA molecule

 New Methodology: attempt to bind probe/reporter complex with fluorescent marker. If bound, we observer strong fluorescence signal. Interactivity Probe/Reporter complex If probe can bind, we get fluorescence signal, indicating accessibility. If a probe cannot bind, then we get no fluorescence signal, indicating inaccessibility.

Knowledge Gradient scores High confidence in prior DMS data Each bar is a potential region to probe. The vertical axis is the KG score.

Knowledge Gradient scores Low confidence in prior DMS data This plot tells us how much information can be gained from targeting each region. Highest scoring regions = most information to be gained

Knowledge Gradient scores Use KG scores as a guideline to picking the next experiment. The primer for this highest scoring probe need to be ordered But we have this primer in stock, and it has a reasonably large KG value.

DM US Hex NB La Temperature (C) Concentration Phase Diagram

Policy function approximations Lookup table policies arise in many settings in everyday life © 2013 W.B. Powell Slide 11