urol-e5 / timeseries

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

Check Biomass calculations between host and sym #37

Closed hputnam closed 2 years ago

hputnam commented 2 years ago

@AHuffmyer can you check lines 49-51 on Biomass.Rmd to confirm that we have the correct order of the denominator for sym and host.

https://github.com/urol-e5/timeseries/blob/master/timepoint_1/scripts/biomass.Rmd

Data <- Data %>% mutate(dry.pan.mass.g.ml = case_when(partner=="Host" ~ dry.pan.mass.g/sym, partner=="Sym" ~dry.pan.mass.g/host), burnt.pan.mass.g.ml = case_when(partner=="Host" ~ burnt.pan.mass.g/sym, partner=="Sym" ~burnt.pan.mass.g/host))

hputnam commented 2 years ago

It looks like this was done to fix a volume mixup flip flop on lines 32-34. We should fix this at the source (lines 32-34) and in lines 49-51 so it is not confusing

different volumes for sym (4ml) and host (5ml)

sym <- 5 host <- 4

AHuffmyer commented 2 years ago

I see - so are you saying that it be that symbiont = 4 and host = 5 on lines 32-34? Rather than sym = 5 and host = 4?

AHuffmyer commented 2 years ago

We have the same issue for other time points as well, so I can fix all of those accordingly.

hputnam commented 2 years ago

yes

hputnam commented 2 years ago

Also there is an issue with the calculations. I tried to simplify this by accounting for the homogenate volume first. Then subtracting apples from apples. Importantly the DW had dry biomass - burned biomass, and that should be dry biomass - initial pan

AHuffmyer commented 2 years ago

Scripts corrected, re ran to look for missing data. There are several colonies in TP4 that have missing values due to missing homogenate volume and surface area values, need to decide what to do about this. I also corrected several typos/incorrect colony id's in dataframes.

AHuffmyer commented 2 years ago

This also corrected some of the high values for symbiont:host biomass ratios that we had earlier. Ratios are now ~0.5 as compared to 0.8-1.5 in previous calculations. @jrcunning these values seem much closer to what we would expect! Biomass_Figure.pdf - check out the third panel on this plot (ignore the outlier on the middle plot, I'll correct that on my next round of QC).