Matching of graph peaks over time I am a programmer but my data analyses/statics skills are non-existent. I am a quick learner though and no problem I have set to solve has yet to become unattainable (let's hope this is not the first one ;) ). Therefore I was hoping that someone could give me some help as to where start looking.
I have a graph with x,y (chromatography - relative level or time) readings that look similar to the following:

For each peak it will be manually assigned a molecule depending on its retention time (x). By matching the peaks from the new graph with previous ones we would by consequence find also the assigned molecule. What would be the best approach in order to apply that assigned detection to a new reading by comparing it with previous readings, having in consideration that the peak levels can vary, a peak can also happen to not exist in a reading, there might be some noise in the readings (much smaller that are not peaks), and the x can slightly shift?
 A: While I don't have experience with chromatographic data and its specialties, and with the exception of Ron Wehrens' book I don't know the papers I linked below. However, here are the results of a quick literature search that I'd use to get started on the subject:


*

*Michał Daszykowski and Beata Walczak have a paper in TrAC on Use and abuse of chemometrics in chromatography.
As it is in TrAC I expect it to be a good overview article.

*There's a brand new book on Chemometrics in Chromatography

*Ron Wehrens' Chemometrics with R has a section on parametric/dynamic time warping (pre-processing techniques that align the same peaks to correct for changes in retention time due to aging of the columns).
There are also tons of papers out on the topic - you'll probably have to try what works well with your data (extrapolated from my experience with pre-processing spectroscopic data).





*

*I notice that people use PARAFAC2, a 3-way model (in chemometric terminology: tri-linear model) that allows unequal axes such as time shifts/warping in one dimension. 

