The wonders of science
One of the frustrating things about science is trying to repeat results - although as this entry will demonstrate it's a good job scientists make sure their results are repeatable.
My "big" project at the moment is related to a protein that appears to cut another protein in half and that this somehow makes fly wings (and other parts of it) rounder and fatter. It hasn't been shown to make people rounder and fatter though, although it's levels are known to be altered in some cancers. As an aside I sometimes think a more grounbreaking result would be to find a gene whose activity remains completely unaltered in all forms of cancer - the biological equivalent of the the luminiferous aether.
Anyhow 18 months ago I was able to show via Western Blots that RNAi (a way of reducing gene activity) of my favourite gene caused a reduction in the other protein being cut. I managed to repeat this 3 times and thought "jobs a good un". A few months ago I made a mutant of the gene which in theory should behave just like the RNAi line. Well it doesn't. At best there appears to be more of the other protein (both cut and uncut). This is a bit worrying so i went back to the RNAi lines to see whether I still get the same result. In this particular experiment I managed to get an increase and a decrease in two samples which should have been essentially the same. All the controls suggest the experiment makes "sense". Have I inadvertently stumbled upon Schrodinger's protein - something which can increase and decrease it's activity at the same time? I doubt it, so I shall resort to what most scientists do when hit over the head with baffling results - assume I fucked up somewhere and repeat the experiment.
Another valuable lesson in all of this is that experiments you are planning on putting into an upcoming talk invariably work or at worst succeed in making you doubt what you thought was real. Seeming though I'm giving my "farewell" seminar to the entire department very shortly I will be applying a significant amount of spin on recent events.
My "big" project at the moment is related to a protein that appears to cut another protein in half and that this somehow makes fly wings (and other parts of it) rounder and fatter. It hasn't been shown to make people rounder and fatter though, although it's levels are known to be altered in some cancers. As an aside I sometimes think a more grounbreaking result would be to find a gene whose activity remains completely unaltered in all forms of cancer - the biological equivalent of the the luminiferous aether.
Anyhow 18 months ago I was able to show via Western Blots that RNAi (a way of reducing gene activity) of my favourite gene caused a reduction in the other protein being cut. I managed to repeat this 3 times and thought "jobs a good un". A few months ago I made a mutant of the gene which in theory should behave just like the RNAi line. Well it doesn't. At best there appears to be more of the other protein (both cut and uncut). This is a bit worrying so i went back to the RNAi lines to see whether I still get the same result. In this particular experiment I managed to get an increase and a decrease in two samples which should have been essentially the same. All the controls suggest the experiment makes "sense". Have I inadvertently stumbled upon Schrodinger's protein - something which can increase and decrease it's activity at the same time? I doubt it, so I shall resort to what most scientists do when hit over the head with baffling results - assume I fucked up somewhere and repeat the experiment.
Another valuable lesson in all of this is that experiments you are planning on putting into an upcoming talk invariably work or at worst succeed in making you doubt what you thought was real. Seeming though I'm giving my "farewell" seminar to the entire department very shortly I will be applying a significant amount of spin on recent events.