welcome to our quantification with drop the digital pcr instructional module my name is frank biswarn and i am part of the digital biology group of biorad laboratories in this short module i will introduce the concept of digital pcr and describe how we can accurately count nucleic acid molecules at some point in your molecular biology work you surely ask the question how many copies of my target molecule are present in my sample chances are you have used the most popular technique for quantifying nucleic acids quantitative pcr quantitative pcr or qpcr is a powerful technique but it has its limitations when higher accuracy and precision is required or when the sample or targets are more challenging droplet digital pcr is the go-to technology you have probably heard of ddpcr as you are currently watching this video what you may not know is that digital pcr was invented before qpcr but it was unaffordable and technically challenging until recently ddpcr is a direct molecule counting technique it can be used to quantify both dna and rna as you can see the results can be quite impressive this is an example of a duplex reaction where a 10 dilution series of staph aureus was amplified in the presence of a constant concentration of human genomic dna the error bars represent 95 confidence intervals you may be wondering how these levels of precision are achieved when quantifying a target nucleic acid it boils down to a combination of chemistry and statistics on the chemistry side of things ddpcr involves taking what would be a typical 20 microliter qpcr reaction and breaking it up into tens of thousands of sub-reactions in a process called partitioning these sub-reactions are then amplified in a thermal cycler and subsequently analyzed for the presence or absence of the target of interest finally the data points are plotted on a graph this is an example of what the data for each reaction can look like for each well the amplitude of each droplet is plotted in a graph with fluorescence amplitude on the y-axis and droplet number or event count on the x-axis droplets that had the target of interest in them at the beginning of the reaction had that target region amplified and with the use of probes or dyes generated a subsequent specific fluorescence signal threshold is set to separate what we decide are positive droplets depicted by the blue dots versus negative droplets depicted by the black dots in the example we have here we would set the threshold at about a thousand fluorescent units as ddpcr is an endpoint analysis reaction the actual amplitude of each droplet is not important anything that is reasonably higher than the negative group is assigned as positive these reactions are called digital due to the nature of the analysis with droplets either positive or negative and when translated into computer jargon having a value of one for positive droplets and zero for negative droplets as we count the number of positives it is tempting to simply use that number and call the sample in the example on the screen right now using 6898 positive droplets as the concentration of our target of interest would actually be incorrect to illustrate how molecules are counted here is a simplified view of the process let's pretend we have four samples and we split our reactions into 143 droplets for each sample in the first sample after amplification we do not detect any positive events it is therefore reasonable to assume that sample one is negative for our target of interest in the second sample we detect six positive droplets therefore we can assume that the number of copies of our target of interest is around six in the third sample we detect 34 positive droplets and are probably tempted to say that the number of copies in that sample is 34. in the fourth sample we see 70 positive droplets and here again we are probably tempted to say that we have 70 targets of interest in that sample if we were to use direct counts our analysis for samples sample 2 is a little bit off while samples 3 and 4 would be incorrect this is the point where statistics comes in when molecules are randomly distributed in a set of droplets if we have low levels of our target of interest it's probable that each one will land in its own droplet on the other hand as concentration increases the probability that two of our molecules of interest or more end up in the same droplet increases to correct for this effect we use poisson's equations for his law of small numbers his equations have been around for over 200 years and are very well established this is the formula we use to calculate concentration copies per microliter of reaction equals the negative lawn of the number of negative droplets divided by the number of total droplets this value is then divided by the volume of our droplets with this in mind if we go back to our examples from a few slides ago and look at sample one it would be safe to say that that sample is negative for our target of interest applying poisson correction the other three samples would slightly change sample 2 would change to 6.2 copies instead of the six copies we predicted that is close but the point two difference is for accuracy and is due to the fact that even though we only have six positive events there is a small probability that two of our target molecules may have ended up in the same droplet for the same reason sample 3 is likely to contain 38 molecules of interest instead of 34 and sample 4 96 instead of the 70. as concentration increases so does co-occupancy and these must be corrected using poisson's equation now of course remember that 143 droplets is not a realistic number of droplets to use as with any statistical analysis the more events that are analyzed the more precise and accurate the results will be we normally use at least 10 000 droplets to get good statistical representation of the true value in our sample let's go back to our first example where we saw 6898 positive droplets and 13 232 negative droplets we can calculate the actual concentration by using poisson's equation the negative lawn of negative droplets 13 232 divided by the total number of droplets red 20 130 and then divided by the volume of our droplets which in this example is 0.85 nanoliters would give us a concentration of 494 copies per microliter we would multiply this value by the number of microliters in our reaction in our case 20 and get a total value of 9880 copies for the entire reaction this value is different than the 6898 positive events we have read because it now takes into consideration co-occupancy and the random distribution as we corrected by poisson analysis now of course bear in mind that there is analysis software that will automatically calculate these values for you as you have hopefully seen molecular counting with ddpcr is relatively straightforward and conveys many advantages over qpcr and other nucleic acid quantification techniques precision inaccuracy absolute quantification without standards higher order mult multiplexing high sensitivity and complex backgrounds and robustness all make drop a digital pcr an ideal technology for nucleic acid analysis thank you for taking the time to view this presentation for more information on the digital pcr and its various applications download the drop a digital pcr applications guide bulletin 6407 from biorad.com and or partake in another module from our master classes site additionally a series of overviews and application specific protocols relating to drop a digital pcr is available from springer publishing look for their methods in molecular biology series volume 1768 thanks again you