Rich Hoff Immersion Week 3

This week I was able to make progress related to troubleshooting the TINY. It's strange that these consistent false positives in amplification have only really been observed as of a few months ago and mainly from the negative controls. The LAMP reaction can be finnicky, demonstrating issues with repeatability of quantification for the same sample and also limitations in quantifying lower concentrations of nucleic acids; these limitations have already been previously observed during the TINY's early development. After discussing the negative control amplification further, I acquired the full sequences of all 6 primers used for the TINY's LAMP reaction from the paper I shared in my Week 1 blog post and compared them to the sequences on the vials in the Cesarman Lab that contained the primers. Seeing that the primer sequences matched, I went ahead used a multiple primer analyzer available from Thermo Fisher Scientific (https://www.thermofisher.com/us/en/home/brands/thermo-scientific/molecular-biology/molecular-biology-learning-center/molecular-biology-resource-library/thermo-scientific-web-tools/multiple-primer-analyzer.html) to determine if any primer dimers could be generated. In the results, I observed that a few different cross-dimers could be generated from different pairs of these primers as well as some self-dimers with both primers having the same sequence. While this information is interesting and potentially insightful, it doesn't show the probability of dimer formation nor guarantee that these dimers will form during the reaction. We would have to run gel electrophoresis of the LAMP reaction product in order to confirm the presence of primer dimers. I asked if it could have something to do with the freshness of the master mix containing the primers, and it seems that the Cesarman Lab already attempted to look into this sometime over the past few months and freshness seemed not to be a factor here, at least within the timeframes of master mix preparation and running LAMP reactions that were of interest for the TINY. There has also been talk about the master mix being jostled during shipment, and perhaps an experiment could be ran in the future to see how continued shaking affects the quality of the results and whether that increases the likelihood of primer dimer formation and/or negative control amplification. Contamination also seems to have been ruled out for the most part, with the Cesarman Lab having previously done thorough cleaning of their pipets and such with bleach as a means of eliminating any potential contamination. A closed tube reaction is performed with the TINY and sterile technique is prioritized in the Cesarman Lab, further reducing the possibility of contamination. The TINY can be cleaned with UV light, so that might be worth attempting in case the specific TINY device at Weill Cornell itself is contaminated.

An interesting experiment that I carried out this week seemed to generate results that suggest that the TINY at Weill Cornell itself is responsible for the seeming negative control amplification. I prepared plates for running LAMP reactions with the qPCR machine that tested multiple controls and master mixes to compare fluorescence readings generated from LAMP reactions that were not carried out in the TINY. We wanted to see if negative controls would appear to amplify under these conditions outside of the TINY. I used a newer and older batch of master mix containing the primers of interest that were prepared and sent from my lab group back in Ithaca as well as a master mix batch prepared by the Cesarman Lab. We used multiple master mixes with different ages to account for the role that master mix freshness could potentially have. The controls included a positive control often used for LAMP and qPCR reactions for the Kaposi sarcoma detection (KS-Detect) project, and my two negative controls were salmon sperm DNA (previously used for the TINY) and another negative control that was recently prepared and shipped from my lab back in Ithaca. We prepared two plates for running the reaction. Both plates' reactions in the PCR machine involved minute-long cycles, which is what is done for the LAMP reactions carried out in the TINY; the difference between how the reactions were carried out for these two plates was how often the fluorescence measurements were recorded. For one plate, a measurement was acquired at the end of each cycle (one measurement every minute), while a measurement was acquired every 30 seconds for the other plate. The results were interesting because the negative controls' data observed from the plate where measurements were acquired every minute had strong resemblance to data acquired from negative controls that underwent a LAMP reaction in the TINY, with this data appearing to suggest that the negative controls were amplifying. The negative controls from the plate that received fluorescence measurements every 30 seconds did not amplify, which is what should be observed. This data is not necessarily 100% conclusive, though I do trust the PCR machine and my pipetting technique. I plated everything and was diligent about avoiding error and inconsistent technique. The technician at the Cesarman Lab whom I've been working with said she was told that the TINY acquires readings every minute; I asked Juan (Boza) from my lab about this (he's done a lot of programming revision and troubleshooting for the TINY) and he said that the TINY acquires fluorescence measurements every 5 seconds. I am wondering if maybe the apparent negative control amplification stems from something being wrong with the detection system and sensors within the specific TINY device being used at Weill Cornell. This would mean that there are no primer dimers amplifying that are contributing to the false positives from these controls; it's just that there is a problem with how the readings are being acquired from negative control LAMP reactions within this specific device, leading to the acquired data falsely suggesting that amplification has occurred. Technique-wise everything about the LAMP reaction appears to be carried out according to the specific protocol that was devised and is done in Ithaca. One main difference process-wise appear to be the time difference between when the master mixes were used for LAMP reactions in Ithaca and NYC following their preparation and shipment from Ithaca, but as I said this was not an issue previously and also not an issue when these reagents were shipped to and utilized for LAMP reactions carried out with the TINY in Africa. The other main difference is the specific TINY device being used, as while the device in Ithaca and the device in NYC should function the same and both were assembled in the Erickson Lab, these problems with negative controls seem to be more consistent with the device in the Cesarman Lab. Our results were presented at the group meeting about the KS-Detect project, and it seems that the current plans are to send the reagents to Africa and continue with diagnostic testing with the TINY there; it seems that the current conclusion is that the issue is just isolated to the specific TINY device at Weill Cornell as I've suggested. With that, it seems that I am done with my troubleshooting for the TINY and will move on to working on something else in the Cesarman Lab, which likely includes continuing to run KS diagnostics of skin biopsy samples from Africa with the qPCR machine. I will probably ask Dr. Cesarman about doing some image analysis as she suggested and take up her offer to arrange from me to shadow some clinicians in the pathology department since I haven't had any clinical experience yet or observed any surgeries since I've been in NYC. I will probably also reach out to some friends about joining them in shadowing their clinician mentors sometime during what remains of Immersion. I'm looking forward to that among other things during the second half of Immersion.