Hi. I am a graduate student using Doric Neuroscience Studio.
Since there seems to be not much references about using Doric's fiber photometry console(FPC) (v. 4.x or 5.0.0) and our lab doesn't have any experience in conducting fiber photometry, I would appreciate if anyone can help me on these following questions regarding some basic knowledges about fiber photometry.
1. According to the references, why do we commonly use 531Hz & 211Hz for delivering violet and blue light respectively? What is special about 531Hz & 211Hz, and how are they calculated?
2. When we deliver pulses, should the pulse forms be sine-wave, or can it be on/off square pulse (TTL)?
3. What does it exactly mean to demodulate the signals? What is the purpose? What should be the result of demodulation? (In general, how does the data look like before and after demodulation?)
I'm not an expert, but I have recently investigated fiber photometry systems. I found two papers describing open-source systems to be particularly helpful:
Owen and Kreitzer, J. Neurosci. Methods (2018)
Akam and Walton, Sci. Reports (2019)
What I took from those papers was:
1. There's nothing special about 531 Hz and 211 Hz. You could, if you wished, choose other frequencies. For example, Owen and Kreitzer have this to say about why they chose 217 Hz and 319 Hz for use with GCamp (isosbestic and calcium-dependent signals, respectively):
"Importantly, each oscillation frequency must be rapid relative to the timescale of the signal fluctuations to be detected, but slower than the detection speed of the photodiode and digitizer. In our hands, oscillation frequencies in the range of 100-400 Hz meet these criteria. To avoid generating harmonic interference, each frequency must not be a multiple (or near multiple) of the other frequency, or of any oscillating signal in the environment, such as room lights (60 Hz) or optogenetic stimulation pulse trains. Based on these criteria and empirical testing, we selected 217 Hz as the oscillation frequency for the 465 nm calcium-dependent fluorescence channel, and 319 Hz as the frequency for the 405 nm isosbestic signal."
To this I would simply add that you also want the two frequencies to be far apart from each other (as well as incommensurate) to make separation cleaner.
2. You could use either sine waves (as Owen and Kreitzer do) or on/off signals (as Akam and Walton do). The post-processing steps are different -- Fourier decomposition in the first case but not the second -- but the two methods both seem good. Akam and Walton have some discussion about how they differ; I recommend you read that. I think Doric Neuroscience Studio -- though, I'm not sure about this -- allows you to choose one or the other.
3. In the sine-wave method, the raw signal is modulated at the stimulus frequency. See Fig. 3 of Owen and Kreitzer for an example. Demodulation means using a sliding window and Fourier analysis to pick out the power near the frequency of interest. Fluctuations of the power for the calcium-dependent signal frequency are a measure of the calcium dynamics.
- Badges
- Science method