Tracks/Measurement

From 2014.igem.org

Revision as of 22:49, 29 December 2013 by Jakebeal (Talk | contribs)


iGEM 2014 Measurement New Track

Introduction

Precise measurements lie at the foundation of every scientific discipline, including synthetic biology. The limits of our knowledge are set by how well we can connect observations to reproducible quantities that give insight. Measurement is also an act of communication, allowing researchers to make meaningful comparisons between their observations. The science and technology of measurement are easily overlooked, because measuring devices are so familiar to us, but behind even the simplest devices lies an elaborate infrastructure. Consider a laboratory pipette. How accurate are the volumes it dispenses? How similar is it to other pipettes? How do you know? The answers to these questions are a complex story involving everything from the speed to light in vacuum to the atomic properties of cesium.

In synthetic biology, measurement is a critical challenge that is receiving an increasing amount of attention each year. For example, one of the long-standing goals of both iGEM and synthetic biology at large, is to characterize biological parts, so that they can be more easily used for designing new systems. The aim of the iGEM Measurement Track is to get students informed and excited about these problems, and to highlight the successes that teams are able to achieve in the area of measurement. The Measurement Track also aims to find out what measurement assays teams have available and to lay groundwork for future more complex measurement activities in iGEM.

Measurement Challenges in Synthetic Biology

With all the instruments in our laboratories, why isn't measurement a solved problem in synthetic biology?  Part of the problem is knowing what to measure and in what context.  One way to think about the impact of measurements is in terms of four levels, each building upon the last:

  1. Measurement quantifies a phenomenon that has been experimentally observed.
  2. Quantitative measurements may be used to create a model of how the phenomenon was produced.
  3. Models may be applied to predict what quantitative phenomena will be observed in a new context.
  4. Predictions may be used to inform choices about how to engineer towards desired phenomena.
Instruments, by themselves, only address the first level.  In synthetic biology, many models are constructed, often post-facto. Quantitative predictions, however, are still extremely difficult: an important part of the problem is determining how measurement relates to context, so that we can understand what sorts of things a model can be reasonably expected to predict.

Even when we know what we wish to quantify, it may be impractical to obtain with our current instruments.  For example, many quantitative models describe how the concentration of chemicals in a single cell changes over time.  Behaviors often vary greatly from cell to cell, so it is often desirable to collect data from a large number of individual cells.  Most current instruments, however, cannot readily measure this.  Instead we end up having to make tradeoffs like these:


A mass spectrometer can measure the amount of particular chemicals in a sample, but any cell measured is destroyed, it is difficult to obtain measurement from individual cells, and often difficult to interpret the massive pattern of data produced to quantify particular chemicals of interest. 

A flow cytometer can take vast numbers of individual cell measuremements, but the measurements are of a proxy fluorescent protein rather than the actual chemical of interest and the cells may still be disrupted by running them through the instrument.  Unless calibration controls are run with an experiment, the measurements are relative and non-reproducible.

A fluorimeter is less invasive than a flow cytometer and can measure changing fluorescence over time with little impact on the cells, but still uses a fluorescent proxy.  Its measurements are also of the whole sample rather than individual cells, and also relative to the number of cells in the sample.

A microscope can track and quantify fluorescence from individual cells, but not very many of them, and often needs human help on tracking.
Figure 1: No generally available instrument can measure chemical concentrations in large number of single cells over time.

Relative measurements are a major problem, because they cannot be compared.  If you build models of biological devices using different relative measurements, then you cannot combine the models to predict what will happen when you combine the devices.  If units are relative to a batch of samples or to a laboratory, then you cannot reproduce experimental results: even if two experiments produce the same numbers in a new experiment, if the units are relative you cannot tell whether the results are actually the same or whether they have been uniformly shifted (which might be very important!).

Figure 2: Models using different relative units cannot be compared or connected.  How many "Blue" in the output characterized for Repressor #1 are equal to a "Red" in the input characterized for Repressor #2?

Beyond these core scientific concerns, there are pragmatic problems as well. Instruments are also often very expensive to buy and to operate.  This is an especially big problem for DIY groups and researchers in smaller institutions or developing nations.  Cheaper instruments are sometimes available, but usually produce much less accurate or precise data.  Once you've got the data, you also need to be able to share it effectively, so that everybody can benefit from the information that is being learned.  The community will thus likely also need new tools and data exchange standards to allow for simpler and more effective sharing of measurements and models.

The challenges of measurement in synthetic biology are large and broad.  They cover everything from fundamental biological questions to the need for better cheaper instruments and community data sharing.  But because measurement affects to many things, improvements in any of these areas are likely to have a big impact.

Plans for the Measurement Track in 2014

The 2014 event expands on iGEM's long-running inclusion of measurement as a focus area (a measurement award has been given since 2006).  This year we are introducing a medal for measurement, and splitting the single prior award into two awards.

Teams participating in the Measurement Track in 2014 can earn a medal by taking part in a group measurement project, in which each team measures the same properties of several known samples.  We will provide some recommendations for experimental and measurement protocols, but teams are encouraged to use whatever approach will provide the most reliable and accurate measurements with the resources available to them.  All of the results will be collected together and later shared, which will allow people to see the tradeoffs between different approaches.

Teams will also be able to compete for two awards:

  • Best Characterization Project focuses on performing measurement.  It will be awarded to the team that best gathers high-quality data about the behavior of biological devices, such that this data can aid future engineering projects.
  • Best Innovation in Measurement focuses on improving measurement.  It will be awarded to the team that provides the best improvement to quality and/or accessibility of measurement techniques.

Measurement Track Committee

We have a great committee to help coordinate the Measurement track in 2014.

  1. Chair: Jacob Beal, Raytheon BBN Technologies
  2. Traci Haddock, Boston University
  3. Jim Hollenhorst, Agilent Technologies