nolanlab / citrus

Citrus Development Code
GNU General Public License v3.0
31 stars 20 forks source link

Assumption on channel names #73

Closed shenorr closed 9 years ago

shenorr commented 10 years ago

Hi Rob, In the function .getDisplayNames you make an assumption that channel names would be longer than 2 characters (in the line displayNames[nchar(reagentNames)>2] = reagentNames[nchar(reagentNames)>2] ) - This assumption is not documented anywhere. When is this an issue: Channel names are output by the Fluidigm software and are all longer than 2 characters, but for reagents that's not necessarily the case. If users actually have a channel with a reagent that's shorter (as we did :-( ) there will be a mismatch in the assignment above which will yield a warning but code will seem to run fine. Channel names in all plots though will be shifted making things confusing. I fixed it in my own version with a >= :-), but I am unclear why you even have this statement at all. thanks, Shai

rbruggner commented 10 years ago

Hi Shai,

There was a bug in the get display names function when you branched shai4real from master. If you pull the latest changes from master, does that fix the problem?

Rob

On Sep 27, 2014, at 12:33 AM, shenorr notifications@github.com wrote:

Hi Rob, In the function .getDisplayNames you make an assumption that channel names would be longer than 2 characters (in the line displayNames[nchar(reagentNames)>2] = reagentNames[nchar(reagentNames)>2] ) - This assumption is not documented anywhere. When is this an issue: Channel names are output by the Fluidigm software and are all longer than 2 characters, but for reagents that's not necessarily the case. If users actually have a channel with a reagent that's shorter (as we did :-( ) there will be a mismatch in the assignment above which will yield a warning but code will seem to run fine. Channel names in all plots though will be shifted making things confusing. I fixed it in my own version with a >= :-), but I am unclear why you even have this statement at all. thanks, Shai

— Reply to this email directly or view it on GitHub.