Day 2 of TMM2013 began with a video link shared session between this meeting here in Banff and another happening in Leipzig. It was a pretty cool idea and underscores two neat things about the Canadian Stem Cell Network and CCRM. First, despite being viewed as a national stem cell network meeting there is clear emphasis on international collaboration, and second — in another nod to professional development — the SCN/CCRM sent five trainees to Leipzig and hosted five European trainees at this meeting.
As many readers will know, one of the areas of stem cell biology that is nearest and dearest to my heart is single cell analysis. One of the international speakers, Professor Sui Huang from the Systems Biology Institute in Seattle gave an excellent presentation on the multiple dimensions of individual cell state. Huang insisted that even in what we regard as the most homogenous cell populations, there is subtle heterogeneity in cell state and this has consequences on the transition from one cell state to another. He pitched two important concepts that have substantial implications for all biologists concerned with cell differentiation and reprogramming.
The first is borrowed from mathematical theory, and it is the idea that within any population there exist some units that will behave opposite to the expected outcome. For cell biologists that means that despite stimulating cells to go in a particular direction, some will actually do the opposite. He dubbed these cells “rebellious” in that they refuse to behave as directed.
The second is that cells exist in a continuum with cells more or less close to the cell state that is being encouraged. This means that cells on the “close” side of the continuum will transit cell state first and those on the “far” side will do so more slowly (or not at all). These are different to “rebel” cells in that they are moving in the same direction, just at a different rate due to the cell state that they started. To exemplify this, he catalogued work in a reasonably homogenous cell line that showed outlier cells in a first stimulus (i.e., those that do not transit cell state upon stimulation) were more likely to change states following a second stimulation, suggesting that they were on their way, but were living too far away from the transition point when first stimulated.
So, what does this mean for experimentalists moving forward? Much work in recent years has focused on single cell analysis and Huang insisted that such work needs to be undertaken with incredible care. The difficulty is that when you assay a single cell from a population, you only get one shot – this means that it is difficult to distinguish outliers and rebels from technical noise (i.e., an assay failure). An example that Huang provided was in single cell PCR experiments where his group did 100 repeats of the same reaction and showed that for lowly expressed genes, the variability was large and good confidence in the data could only be determined with multiple repeats.
The question/answer period was appropriately stimulating with Fabio Rossi firing off a series of questions about potential sources of variability prior to the technical replicates. Controlling for such changes prior to amplification is virtually impossible though so these assays will always be a little dirty.
Overall, this means that 100s of cells need to be assayed in order to determine the behaviour of any given population. This isn’t so bad when it comes to PCR experiments, but when one thinks about extending these concepts to functional assays, the challenges (and budget line) grow substantially. Bottom line, be careful how you interpret outliers and rebels – they might not be what you think.
Latest posts by David Kent (see all)
- Steady progress and more interesting science – 10 years of iPS cells - August 25, 2016
- Tip of the iceberg? Scary stuff from international stem cell clinics - August 2, 2016
- ISSCR final day – Not just one hit… finding combinations in cancer - July 18, 2016