My project seemed very simple in the beginning: no difficult biochemistry, no complicated experimental procedures, and honestly, the fanciest machine ever involved in my experiments is probably the thermal cycler, aka the PCR machine. But in reality, as I’ve slowly found out, problems never stop, frustrations never stop. They come and go one after another, and I not only have to solve them scientifically, but also have to deal with them mentally.
The first real problem occurred as early as I was testing the primers I designed to amplify the gene of interest. I designed 4 different primer sets to amplify 4 different portions on the gene. Despite that theoretically, they should all work perfectly, oftentimes the case was, either the PCR reaction didn’t work at all and I didn’t see my target on the gel, or multiple nonspecific amplifications occurred and lots of bands of the wrong size showed up on the gel. I spent almost the entire first half of the program perfecting PCR — redesigning primers, playing with the annealing temperature of each primer, adjusting the amount of reagents in the reaction, etc. Everyday just involved an enormous amount of PCR and gel electrophoresis and attention to details: keeping detailed notes of what changed and what remained unchanged, while making sure I didn’t use up too much of a patient’s DNA for testing and troubleshooting.
When I got my PCRs working after almost 4 weeks of troubleshooting, I could finally send the products to sequencing. My hopes were high: after the sequences are back, I could start searching for mutations! But again, the sequencing reactions didn’t go smoothly. For mutation-searching, the quality of the signals is very important. Any contamination could bias the result, but the sequences were seldom free of contamination/noise. As a result, I had to start troubleshooting the sequencing reaction: designing new primers for sequencing, adjusting the concentration of the template, etc. As of today, there are still a few samples of which I haven’t got perfect sequences, and I’m still working hard on perfecting the results.
For the samples whose sequences were clean enough, I started to actually search for mutations. Mutations are, however, really, really, rare. So far, I’ve only spotted three, among which two are synonymous (they don’t cause a change in amino acid) and therefore clinically irrelevant. The most exciting one showed up just two days ago because it actually caused an amino acid change from tyrosine to cysteine. My mentor and I spent the entire afternoon almost frantically looking up this mutation, trying to figure out the details of it such as its functional relevance. Disappointingly however, it turned out that it was just a common variant across human population: it has an allele frequency of 1.8%, which is way too high to be considered rare or pathogenic.
In summary, I’ve not gotten any real “results” in my project. But in retrospect, I actually enjoyed this whole troubleshooting->working->failing process rather than got beaten by it. What’s exciting about scientific research is not entirely the groundbreaking discoveries, but all the different paths that lead to that destination as well as the views along them. I will continue working hard on my project and try my best to search for mutations, even if they are not there.
