[aroma.affymetrix] Re: ufl and ugp files for na27?
Dear Henrik Thank you very much for your reply and your scripts. First, the differences between na26 and na27 are as follows: Header: - line 11: changed annotation date format from July 21 2008 to 2008-12-01 Data: - line 17: added one new column, i.e. last column is now %GC - the data for Probe Set ID,Chromosome,Physical Position are identical for na26 and na27. This means that in principle I can still use na26. Second, when I compare your results for HindIII, XbaI and NspI for na26 to my results with na26 and na27, I get identical summary results for both ufl and ugp files,as you may have realized when adding your results for na26. In contrast, for StyI I my results for both na26 and na27 are still: snp cnp affxSnp other total enzyme1-only 144868 0 0 0 144868 missing 93436 0 074 93510 total238304 0 074 238378 This means I get 93436 missing snps vs your 607 missing snps. (See the partial output for importFrom below!!!) Even downloading the na26 file and the cdf file from Affymetrix again did not change the result, which is strange since for the other 3 chiptypes the results agree with your results. BTW, I have downloaded 250k_sty_libraryfile_rev4.zip. However, there exists also an older version 250k_sty_libraryfile_rev3.zip. Which version have you used for creating the ufl file? Third, trying to run your script Mapping250K_Sty,UFL,na26.R gave the following error at line units - importFrom(ufl, csv, enzymes=enzyme, verbose=log); Error in list(importFrom(ufl, csv, enzymes = enzyme, verbose = -50) = environment, : [2009-01-15 12:33:30] Exception: Argument 'enzymes' contains 1 NA value (s). at throw(Exception(...)) at throw.default(sprintf(Argument '%s' contains %d NA value (s)., .name, sum( at throw(sprintf(Argument '%s' contains %d NA value(s)., .name, sum (is.na(x) at getNumerics.Arguments(static, ..., asMode = integer, disallow = disallow) at getNumerics(static, ..., asMode = integer, disallow = disallow) at getIntegers.Arguments(static, ..., range = range) at getIntegers(static, ..., range = range) at method(static, ...) at Arguments$getIndices(enzymes, range = c(1, 10)) at readDataUnitFragmentLength.AffymetrixNetAffxCsvFile(csv, enzymes = enzymes, at readDataUnitFragmentLength(csv, enzymes = enzymes, rows = keep, ..., verbos at importFromAffymetrixNetAffxCsvFile.AromaUflFile(this, src, ...) at importFromAffymetrixNetAffxCsvFile(this, src, ...) at importFrom.AromaUnitTabularBinaryFile(ufl, csv, enzymes = enzyme, verbose = at importFrom(ufl, csv, In addition: Warning message: In eval(expr, envir, enclos) : NAs introduced by coercion Importing (unit name, fragment length+) data from AffymetrixNetAffxCsvFile...done This is strange since I have tested that argument enzymes returns correctly StyI. Maybe this has to do with my version of aroma.affymetrix_0.9.4. Here are the partial outputs for importFrom(ufl, csv, verbose=-50); for NspI and StyI: 1. Running importFrom(ufl, csv, verbose=-50) for NspI has the following partial output (see ==): Reading AffymetrixNetAffxCsvFile...done Extracting fragment lengths from ([enzyme], lengths, start, stop)... Inferring if enzyme names are specified... Has enzyme names: TRUE Inferring if enzyme names are specified...done Identifying number of enzymes... nbrOfEnzymes 1 2 3 ==131564 130698 2 Max number of enzymes: 3 Identifying number of enzymes...done Splitting into subunits and padding with NAs... Splitting into subunits and padding with NAs...done Extracting enzyme names... Identified enzymes: NspI, StyI Extracting enzyme names...done Identifying the location of the fragment lengths... Offset: 3 Identifying the location of the fragment lengths...done Extracting fragment lengths... Summary of *all* fragment lengths: Min. 1st Qu. Median Mean 3rd Qu. Max. NA's 19.0526.0755.0842.9 1049.0 2000.0 394527.0 Extracting fragment lengths...done Sorting data by enzyme... V1 V2 V3 Min. : 100.0 Min. :19 Min. :NA 1st Qu.: 476.0 1st Qu.: 937 1st Qu.:NA Median : 644.0 Median : 1280 Median :NA Mean : 648.9 Mean : 1231 Mean : NaN 3rd Qu.: 816.0 3rd Qu.: 1614 3rd Qu.:NA Max. :1480.0 Max. : 2000 Max. :NA == NA's : 701.0 NA's :131564 NA's :262264 Sorting data by enzyme...done int [1:262264, 1] 574 700 580 631 666 1060 798 842 822 608 ... Extracting fragment lengths from ([enzyme], lengths, start, stop)...done Reading (unitName, fragmentLength+) from file...done 2. Running importFrom(ufl, csv, verbose=-50) for StyI has the following partial output: Reading AffymetrixNetAffxCsvFile...done Extracting fragment lengths from ([enzyme], lengths, start, stop)... Inferring
[aroma.affymetrix] Re: ufl and ugp files for na27?
Hi Christian,
I'm quite swamped but I've added links the scripts that I used to
generate the existing NetAffx 26 (na26) UFL and UGP to each of the
different SNP CN chip type pages, e.g.
http://groups.google.com/group/aroma-affymetrix/web/mapping250k-nsp-mapping250k-sty
More comments below.
On Mon, Jan 12, 2009 at 6:24 AM, cstratowa
christian.strat...@vie.boehringer-ingelheim.com wrote:
Dear Henrik
Meanwhile I have created ufl and ugp files for both 100K and 500K
arrays but not for GenomeWideSNP_6 aray.
The above scripts will show you how to do it for GenomeWideSNP_6.
Slightly more complicated since two NetAffx CSV files are involved.
Can you confirm that the following code, which I use for both 100K and
500K arrays, is correct:
# retrieving annotation files
chiptypes - c(Mapping50K_Hind240, Mapping50K_Xba240)
cdfs - lapply(chiptypes, FUN=function(x){AffymetrixCdfFile$byChipType
(x)})
names(cdfs) - chiptypes
print(cdfs)
# importing data from NetAffx CSV files
csvs - lapply(cdfs, FUN=function(cdf){AffymetrixNetAffxCsvFile
$byChipType(getChipType(cdf), tags=.na27)})
print(csvs)
# allocating empty UFL (Unit Fragment Length) files
ufls - lapply(cdfs, FUN=function(cdf){AromaUflFile$allocateFromCdf
(cdf, tags=na27,CS20090112)})
print(ufls)
# import SNP data
units - list();
for (chipType in names(ufls)) {
ufl - ufls[[chipType]];
csv - csvs[[chipType]];
units[[chipType]] - importFrom(ufl, csv, verbose=-50);
}
str(units)
# allocating empty UGP (Unit Genome Position) files
ugps - lapply(cdfs, FUN=function(cdf){AromaUgpFile$allocateFromCdf
(cdf, tags=na27,CS20090112)})
print(ugps)
# import SNP data
units - list();
for (chipType in names(ugps)) {
ugp - ugps[[chipType]];
csv - csvs[[chipType]];
units[[chipType]] - importFrom(ugp, csv, verbose=-50);
}
str(units)
This looks alright to me. You might want to check toward the posted
NA26 scripts as well, because they are more recent.
Here is the summary for the 100K arrays:
# Summary 50K chips
str(units)
List of 2
$ Mapping50K_Hind240: int [1:57244] 18632 18677 1631 18713 1630 18712
18619 1639 18722 18608 ...
$ Mapping50K_Xba240 : int [1:58960] 29181 18239 31302 19831 47750
45114 19103 39711 19772 37811 ...
ufl - AromaUflFile$byChipType(chiptypes[1], tags=na27,CS20090112);
print(summaryOfUnits(ufl, enzymes=HindIII))
snp cnp affxSnp other total
enzyme1-only 56933 0 0 0 56933
missing311 0 055 366
total57244 0 055 57299
The NA26 version gives:
snp cnp affxSnp other total
enzyme1-only 56933 0 0 0 56933
missing311 0 055 366
total57244 0 055 57299
ufl - AromaUflFile$byChipType(chiptypes[2], tags=na27,CS20090112);
print(summaryOfUnits(ufl, enzymes=XbaI))
snp cnp affxSnp other total
enzyme1-only 58616 0 0 0 58616
missing344 0 055 399
total58960 0 055 59015
NA26:
snp cnp affxSnp other total
enzyme1-only 58616 0 0 0 58616
missing344 0 055 399
total58960 0 055 59015
ugp - AromaUgpFile$byChipType(chiptypes[1], tags=na27,CS20090112);
print(summary(ugp, enzymes=HindIII))
chromosomeposition
Min. : 1.000 Min. :48603
1st Qu.: 4.000 1st Qu.: 34667112
Median : 7.000 Median : 72677620
Mean : 8.402 Mean : 80405004
3rd Qu.: 12.000 3rd Qu.:114826216
Max. : 23.000 Max. :246727435
NA's :363.000 NA's : 363
NA26:
print(summary(ugp));
chromosomeposition
Min. : 1.000 Min. :48603
1st Qu.: 4.000 1st Qu.: 34667112
Median : 7.000 Median : 72677621
Mean : 8.402 Mean : 80405004
3rd Qu.: 12.000 3rd Qu.:114826216
Max. : 23.000 Max. :246727435
NA's :363.000 NA's : 363
print(table(ugp[,1]));
123456789 10 11 12 13 14 15 16
4541 5072 3962 4342 4215 3968 3444 3549 2357 2743 2466 2592 2661 1931 1440 1145
17 18 19 20 21 22 23
985 1731 326 993 883 433 1157
ugp - AromaUgpFile$byChipType(chiptypes[2], tags=na27,CS20090112);
print(summary(ugp, enzymes=XbaI))
chromosomeposition
Min. : 1.000 Min. :93683
1st Qu.: 4.000 1st Qu.: 34636629
Median : 7.000 Median : 72249739
Mean : 8.507 Mean : 80010574
3rd Qu.: 12.000 3rd Qu.:114666170
Max. : 24.000 Max. :246885089
NA's :390.000 NA's : 390
NA26:
print(summary(ugp));
chromosomeposition
Min. : 1.000 Min. :93683
1st Qu.: 4.000 1st Qu.: 34636629
Median : 7.000 Median : 72249739
Mean : 8.507 Mean : 80010574
3rd Qu.: 12.000 3rd Qu.:114666170
Max. : 24.000 Max. :246885089
NA's :390.000 NA's : 390
print(table(ugp[,1]));
123456789 10 11 12 13
