search

urol-e5 / timeseries

Data generated from e5 time series sampling in Moorea
2 stars 0 forks source link

Timepoint 4 antioxidant capacity standard curve fit #16

Closed AHuffmyer closed 3 years ago

AHuffmyer commented 3 years ago

The standard curve for antioxidant capacity rmd in TP4 has issues in fit - the standard data points are variable and do not lie on the projected curve as well as in other timepoints. See attached photo.

Screen Shot 2021-01-21 at 10 48 28
AHuffmyer commented 3 years ago

Further, TP3 uses a linear model as a fit for the standard curve while TP1 and TP2 use nonlinear models. Which model do we want to use for standard curves in these scripts?

AHuffmyer commented 3 years ago

See example of TP1 standard curve here with expected fit:

Screen Shot 2021-01-21 at 10 51 40
daniellembecker commented 3 years ago

For plates 1 and 2, the standard curve was off because the rxn buffer we used was old. Only reference the standard curve amounts from plate 3 initial and final because those reacted correctly. Going through past data we noticed that the standards were similar throughout all plates so the standard curve from just plate 3 can be applied to plates 1 and 2.

AHuffmyer commented 3 years ago

Thank you, @daniellembecker! This is very helpful. I'll work on this.

AHuffmyer commented 3 years ago

Data using only standard from plate 3 and removing outlier well (b8 with initial value far outside expected). Looks good! Screen Shot 2021-01-21 at 12 53 10