Data repetitions

Those data repetitions are still happening. This is the first place I see but it could be happening in data processing:

Y1_fairbanks/code/fticr-3b-relabund-byhorizon.R

Lines 13 to 26 in 6d6e53f

    
           horizonation_relabund =  
        
             fticr_data_water %>%  
        
             left_join(select(fticr_meta_water, formula, Class), by = "formula") %>%  
        
             ## create a column for group counts 
        
             mutate(Class = factor(Class, levels = c("aliphatic", "unsaturated/lignin", "aromatic", "condensed aromatic"))) %>%  
        
             #group_by(slopepos, cover_type, plot, ID, Class) %>% 
        
             group_by(ID, Class) %>%  
        
             dplyr::summarize(counts = n()) %>%  
        
             ## create a column for total counts 
        
             group_by(ID) %>% 
        
             dplyr::mutate(totalcounts = sum(counts)) %>%  
        
             ungroup() %>%  
        
             mutate(relabund = (counts/totalcounts)*100, 
        
                    relabund = round(relabund, 2))

Sample IDs
3 & 4,
52 & 53,
70 & 71 & 72,
77 & 78

@kaizadp

Is the issue here? We never talked about this part of the code for the Y1 samples. Some plots will have one rep and others will have 3. It doesn't seem like this should be the issue, but since I did it on my own it's worth taking a look at.

Y1_fairbanks/code/fticr-1-dataprocessing.R

Lines 141 to 153 in 5bde8e7

    
           rename(max_reps = reps) %>%  
        
           group_by(slopepos, cover_type, plot, formula) %>%  
        
           dplyr::mutate(formulareps = n()) %>%  
        
           # set up replication filter for 2/3 of max_rep 
        
           ungroup() %>%  
        
           mutate(include = formulareps >= (2/3)*max_reps) %>%  
        
           ## mutate(include = formulareps > 1, 
        
           ##        occurrence = case_when(formulareps == max_reps ~ "3/3", 
        
           ##                               formulareps < max_reps & formulareps >= (2/3)*max_reps ~ "2/3+", 
        
           ##                               formulareps >= (1/3)*max_reps ~ "1/3+", 
        
           ##                               formulareps < (1/3)*max_reps ~ "exclude")) %>%  
        
           filter(include)

Oooookay...I "fixed" it. I may have created more problems. But the data was happening because of the formula reps line. I changed it to the following. It fixed it...there are no more repetitions. I'm going to start redoing figures...hopefully I haven't messed it up even more.

Y1_fairbanks/code/fticr-1-dataprocessing.R

Lines 140 to 153 in 84782ed

    
           group_by(slopepos, cover_type, plot, formula) %>%  
        
           dplyr::mutate(formulareps = n()) %>%  
        
           # set up replication filter for 2/3 of max_rep 
        
           ungroup() %>%  
        
           mutate(include = formulareps >= 1) %>%  
        
           #(mutate(include = formulareps >= 2/3)*max_reps) 
        
           ## mutate(include = formulareps > 1, 
        
           ##        occurrence = case_when(formulareps == max_reps ~ "3/3", 
        
           ##                               formulareps < max_reps & formulareps >= (2/3)*max_reps ~ "2/3+", 
        
           ##                               formulareps >= (1/3)*max_reps ~ "1/3+", 
        
           ##                               formulareps < (1/3)*max_reps ~ "exclude")) %>%  
        
           filter(include)

	horizonation_relabund =
	fticr_data_water %>%
	left_join(select(fticr_meta_water, formula, Class), by = "formula") %>%
	## create a column for group counts
	mutate(Class = factor(Class, levels = c("aliphatic", "unsaturated/lignin", "aromatic", "condensed aromatic"))) %>%
	#group_by(slopepos, cover_type, plot, ID, Class) %>%
	group_by(ID, Class) %>%
	dplyr::summarize(counts = n()) %>%
	## create a column for total counts
	group_by(ID) %>%
	dplyr::mutate(totalcounts = sum(counts)) %>%
	ungroup() %>%
	mutate(relabund = (counts/totalcounts)*100,
	relabund = round(relabund, 2))

	rename(max_reps = reps) %>%
	group_by(slopepos, cover_type, plot, formula) %>%
	dplyr::mutate(formulareps = n()) %>%
	# set up replication filter for 2/3 of max_rep
	ungroup() %>%
	mutate(include = formulareps >= (2/3)*max_reps) %>%

	## mutate(include = formulareps > 1,
	## occurrence = case_when(formulareps == max_reps ~ "3/3",
	## formulareps < max_reps & formulareps >= (2/3)*max_reps ~ "2/3+",
	## formulareps >= (1/3)*max_reps ~ "1/3+",
	## formulareps < (1/3)*max_reps ~ "exclude")) %>%
	filter(include)