 So that is what we call Protein engineering work. That is the multiple most critical process during a sensor development We pick up we pick up those we amplify those genes which encode the GPCRs from our human genome From our human DNAs. We clone those stuff use a technological PCR Polymerase chain reaction. We got the DNA encoding those GPCR and then We can use some enzymes to split it into like two or three different segment and then you got some slot to do and And and you got you like you got like one slot if you cut it You cut it this way you got one slot and then I Clone we clone the fluorescent protein from Jellyfish yes from jellyfish the green fluorescent Proteins from the jellyfish the red fluorescent proteins from the coral So we clone the fluorescent protein from the coral or from the jellyfish and then we got the DNA fragment and We insert the DNA fragment into the GPCR slot and then these three part comes together Was sealed by some enzymes and the way in this way we got the whole DNA sequence for a neural transmitter sensors Actually, you can you can not just do it that way and we got a very perfect sensor with good Expression level with good fluorescence even with good signal to noise ratio. You need to You need to engineer very carefully where to insert the Fluorescent protein into the GPCR backbone and you need to Also engineer the linkers the linker protein or DNA sequence between them that will defines how efficient how how the efficient of the So-called conformational change transfer from GPCR to the fluorescent protein there might be some linkers So you need to tweet you need to do we call it random mutation to those linker sequence to see which amino acid is the perfect and After that, we also engineer some Do some mutations within the GPCR we call the Neurotransmitter is very small that will bind with some specific place within the GPCR We call it binding pocket and within the binding pocket DNA sequence will or Amino assay sequence will defines affinity between sensor or between GPCR with the Neurotransmitter So we also engineer some sort of Specific site within the binding pocket to guard the perfect sensor ways Mmm With good spatial temporal not not with fast kinetics with good signal to noise ratio to transfer the chemical signal into fluorescent signal and Yes, that is a lot of work if you are a First year graduate student just start to do such kind of we call it protein engineering you take like three days or two days like three days to get such a construct and the test it whether it works or not and to cause a perfect Sensor probably you need to try More than one thousand alternatives to got a perfect construct and you need to That that is a lot of work, but but we we figure out some way to do it paralleling So that can save some time, but it's still a lot of work for a graduate student to got such a perfect sensor and Actually, I'm very lucky when I joined the lab We already got those sensors Some are prior students in our lab already build the sensors and test those sensors the most simple Model culture sales artificial culture sales in the medium sales in the medium we tested those sensors and We already know that it works and what so so it turns to me. I I Take the job to create Transgenic animals to test whether they're the sensors Can whether there's this sensor can be used for In vivo studies in intact animals in the living brain there being specific behavior