peperg
commented
3 years ago hey @MeaghanBartley,
Yeah, this is ok. Great that you did look at the warning, they are produced for a reason!.
In this case, what is happening is that there are a few sample/parameter combinations which don't have data (they have an NA).
There are also 7 samples, which have an ammonia concentration of -5. This is impossible and it is likely that it is a code employed by the analytical lab to indicate that the sample was below the detection level, or any other kind of issue with the sample.
Remember what i mentioned about the issues of using numbers to codify missing values notes about the sample??? this is a good example of how when you use these approaches, sometimes you can inadvertently actually use this "code" as a value. This specially problematic when your data makes it to an external database and becomes part of a much larger dataset, where it gets buried. These should have been dealt with by giving those samples an "NA' value and having a column for notes about why a specific value was missing.
What ggplot is trying to tell you is that there were 257 sample sit could not plot:
- 248 because they were NAs
- 7 because of the "-5" Ammonia value. You cannot log a negative value, and because the plot has a logged y scale, these values become NAs as well (good! otherwise we would have plotted negative ammonia values!)
MeaghanBartley
When creating my box plots I received this error message:
NaNs produced Transformation introduced infinite values in continuous y-axis Removed 257 rows containing non-finite values (stat_boxplot).
The box plots themselves look the same as those in the figure we are to replicate. However, I was wondering if there might be something wrong with my code. I figured this might be because some of the nutrients have NAs under their concentration.