Skip to main content ↑ ↓

Show Posts

This section allows you to view all Show Posts made by this member. Note that you can only see Show Posts made in areas you currently have access to.

Messages - panos

1
XCMS / Any possible info about a retcor error?
Dear xcms developers,

I would like to ask (I know it is a quite general question though...) what might be causing the following errors (not at the same time but using different parameters) in a 2nd pass of retcor after a first data grouping:

Code: [Select]
Error in stepfun(rtcor[[i]][-1] - diff(rtcor[[i]])/2, rtcor[[i]] - rtdevsmo[[i]]) : 
  stepfun: 'x' must be ordered increasingly
Code: [Select]
Error in quantile.default(abs(rtdevsmo[[i]]), 0.9) : 
  missing values and NaN's not allowed if 'na.rm' is FALSE
I am trying to process 75 samples of CE/MS data in netcdf format, after filtering out 2 time intervals from the beginning and the end of the scan range. I know that xcms was not initially developed for CE/MS data but it has been sucesfully used in a couple of publications. I should mention that these errors occur in any (steps, snthresh) configuration (matchedFilter), other than the default (steps=1, snthresh=10). For example I have tried (steps=1, snthresh=7), (steps=3, snthresh=10), (steps=3, snthresh=7), (steps=2, snthresh=7) and (steps=2, snthresh=10) is currently running. I am using sigma=7 (found to be good for my data after some visual inspection). The errors are always occuring in an iterative grouping/rt correction procedure, on the 2nd iteration (initial grouping bw=30 or 50, next values 30 or 20). The parameter steps=3 was suggested in a publication using xcms with CE/MS data.
These errors was not occuring in the previous xcms version I had on my system (1.23.1), when using several parameter sets like the above. Thanks in advacne for any replies!

Panos
2
CAMERA / Re: Isotope extraction
Hi Ricca,

Does the following do what you want?  :)

Code: [Select]
# Just a small backup
iso <- isolist
# Remove $charge so as to easily append to data.frame
iso <- lapply(iso,function(x) { x$charge <- NULL; return(x) })
# Create the data.frame
d <- do.call("rbind",lapply(iso,data.frame))
Panos
3
XCMS / Re: Extract metabolite information
Hi Ricca,

I recently started playing with metabolomics MS data and I have run into many of the questions you have. First of all, I think that since you are playing with a list of masses, there isn't really a golden standard rule that can lead you to a definite identification of what a peak represents... What you can do is to start restricting the possible interpretations according to your experiments. For example, regarding the charge state parameters, you could restrict your results according to your instrument/experimental configuration so as to obtain an impression of what kind of adducts you expect to have. E.g., for ESI you could possibly restrict yourself to +/-H adducts.

To obtain as much info as possible for your peak list after xcms, you should use the functionalities of CAMERA package which is really helpful(congrats to the authors once again!) and works smoothly with xcms objects. A further grouping according to the theoretical isotope distribution obtained with CAMERA as well as the adduct list should shed some light on your questions. Using the extra info you could obtain monoisotpic masses for several peaks and from there search databases (METLIN, ChEBI, KEGG etc.). I have tried the above so far with a certain rate of success. Good luck!

Panos
4
XCMS / Re: Difference of minfrac and minsamp in group function
Do these parameters exclude each other? That is the user can use either minfrac either minsamp, depending on whether he/she wants to constrain based on percentage or absolute number of samples in each group? Probably...
5
XCMS / Difference of minfrac and minsamp in group function
Dear xcms developers,

I would be grateful if you could explain the difference between the minfrac and minsamp parameters in the group function of xcms. A small example would also be appreciated. I believe I correctly understand the minfrac parameter (the "percentage" of samples that need to have proper peaks under a group so as to define a meta-peak?), but can you please also explain the minsamp parameter? In addition, if the minsamp parameter is increased, naturally the number of peak groups decreases. How advisable/useful is to increase minsamp in a dataset of 75 samples but including 16 different classes? (CE/MS). Thank you in advance for your help.

Panos
6
METLIN / Re: Install METLIN locally
Hi,

Thanks for the replies. I have looked in the past a little bit on the SOAP interface. It should get part of my job done but it would be nice if you could post at some time an example of how it can be used from R (if it is possible) using e.g. RSOAP. I believe that there are many people out there who would benefit from a more complete programmatic access to METLIN as it represents a great data source... Thanks again.

Panos
7
METLIN / Re: Install METLIN locally
Thanks for the quick reply. Unfortunately, the batch search from the website is not very helpful as long as there isn't any programmatic access to it (e.g. in the form of a PHP query), since I would like to incorporate the search in a pipeline. Or is there some kind of programmatic access? Thanks again!
8
METLIN / Install METLIN locally
Hello,

I was wondering if there is a version of METLIN (even a limited one) that can be downloaded and installed on a local server (using SQL dumps or other flat files), similar to the ChEBI database. As I am trying to build an analysis/annotation pipeline for the lab I work, it would be really useful. Otherwise, is there a programmatic access to METLIN outside the web interface? Thanks in advance!

Panos
9
CAMERA / Re: CAMERA question
Hi all,

I am using CAMERA 1.6.0 on a 64bit Linux machine. The following is part of the result of "cat extended_adducts_pos.csv"

"name","nmol","charge","massdiff","oidscore","quasi","ips"
"[M+H]+",1,1,1.0076,1,1,1
"[M+2H]2+",1,2,2.0152,2,0,0.75
"[M+3H]3+",1,3,3.0228,3,0,0.75
"[M+H+Na]2+",1,2,23.996818,4,0,0.5
"[M+H+K]2+",1,2,39.970758,6,0,0.5
"[M+H+NH4]2+",1,2,19.041423,7,0,0.5
...

I am already using my own ruleset but I would like to ask the following: by looking at the first line, we see that the massdiff for [M+H]+ is 1.0076. Isn't this a bit over-rounding of the difference? As also mentioned above, shouldn't it be 1.0073 (4-digit rounding)? Same for [M+2H]2+ etc. Am I missing something? Thank you in advance!

Panos
10
CAMERA / Re: runParallel slot in xsAnnotate
Thank you Carsten,

While trying to initialize the xsAnnotate objects with multiple cores, I discovered that the "snow" part in the code was missing (CAMERA 1.6.0). There is no problem with this, I added it it myself (the chunk below), but is it supposed to be like this or there is a minor bug there? In addition, are the CAMERA functions parallelized or not yet implemented? I am asking because groupCorr is not really running faster and there isn't any core activity beyond usual. From a fast look at the source they don't seem to be implemented for parallel calculations. Is it so or am I missing something?  :)

The missing part for the initialization with "snow" (taken from the initialization of xcmsSet object):

Code: [Select]
...
else
{
	snow = "snow"
	if (try(require(snow, character.only = TRUE, quietly = TRUE)))
	{
		cat("Starting snow cluster with", nSlaves, "local sockets.n")
		snowclust <- makeCluster(nSlaves, type = "SOCK")
		runParallel <- 1
		parMode <- "SOCK"
	}
}
...

Thank you in advance!

Panos
11
CAMERA / runParallel slot in xsAnnotate
Hi CAMERA developers,

I would like to ask about the purpose of runParallel slot in the xsAnnotate object of CAMERA. As I am processing a rather large dataset, I would like to know if I can take advantage of the mutliple cores of the machine I am working. Thank you in advance!

Panos