[BioC] ADD: Imputation problem in aCGH

[Benjamin Otto]

Hi,

I just found out what the problem is, however a new question arises here. My
chromosomal location provided was in bp not kb. During the imputation the
impute.lowess function segments each chromosome into the two arms using a
included list of centomere locations. At the moment I provide a chromosome
which has only one resident on one of the arms the function moans because it
needs at least two non NA-Values.

Right so this is the old problem. When I had a look at the centromere
locations I found the locations diverging from the positions given by
ensemble. Now how does the function discriminate between the organisms? It
seems that it doesn't and if that is so, then isn't the the separation into
the two arms more a downside than an advantage?


Cheers,

Benjamin


-----Ursprüngliche Nachricht-----
Von: bioconductor-bounces at stat.math.ethz.ch
[mailto:bioconductor-bounces at stat.math.ethz.ch] Im Auftrag von Benjamin Otto
Gesendet: Tuesday, July 17, 2007 11:24 AM
An: bioconductor at stat.math.ethz.ch
Betreff: [BioC] Imputation problem in aCGH

Dear bioconductors,

Trying to impute some log2 ratios with the lowess imputation I get the
following error message, which I can't correlate with my data provided:


# Processing chromosome  2
# Error in approx(lowess(kbl[ind], vecl[ind], f = smooth), xout = kbl[-ind])
: 
#         need at least two non-NA values to interpolate

Now here are two sets of data. The first results in the error message the
second one runs perfectly.

Set #1 (the failing one):
> mx.ratios

              X1874     X2130
AL137761  -0.104402 -0.231907
BC040221   0.520388  0.204723
NM_015677        NA  0.038293
M83653     0.133378        NA
BC033490   0.227578 -0.092427

> mx.info

              Clone Chrom     kb
AL137761   AL137761     2  29114
BC040221   BC040221     2 184720
NM_015677 NM_015677     2 219977
M83653       M83653     2 267639
BC033490   BC033490     2 270090

Set #2 (the working one):

> my.ratios

              X1874     X2130
NM_015677        NA  0.038293
M83653     0.133378        NA
BC033490   0.227578 -0.092427
AK091892  -0.177032 -0.078277
AW451118  -0.092482  0.019423

> my.info

              Clone Chrom     kb
NM_015677 NM_015677     2 219977
M83653       M83653     2 267639
BC033490   BC033490     2 270090
AK091892   AK091892     2 293681
AW451118   AW451118     2 482546


The commands used are:

> mx <- create.aCGH(mx.ratios,mx.info)
> impute.lowess(mx)

Processing chromosome  2
Error in approx(lowess(kbl[ind], vecl[ind], f = smooth), xout = kbl[-ind]) :

        need at least two non-NA values to interpolate



SESSIONINFO:
============
R version 2.5.0 (2007-04-23)
i386-pc-mingw32 

locale:
LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
States.1252;LC_MONETARY=English_United
States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252

attached base packages:
[1] "tools"     "splines"   "stats"     "graphics"  "grDevices" "utils"
"datasets"  "methods"   "base"     

other attached packages:
       aCGH         sma    multtest     Biobase    survival     cluster
hgu133plus2 
   "1.10.0"    "0.5.15"    "1.14.0"    "1.14.0"      "2.31"    "1.11.5"
"1.16.0"



Has anyone an idea what the cause of the error is and why I don't get it in
the second set? I don't see any significant difference...

Best regards,

Benjamin Otto



======================================
Benjamin Otto
University Hospital Hamburg-Eppendorf
Institute For Clinical Chemistry
Martinistr. 52
D-20246 Hamburg

Tel.: +49 40 42803 1908
Fax.: +49 40 42803 4971
======================================



--
Pflichtangaben gemd_ Gesetz |ber elektronische Handelsregister und
Genossenschaftsregister sowie das Unternehmensregister (EHUG):

Universitdtsklinikum Hamburg-Eppendorf
Kvrperschaft des vffentlichen Rechts
Gerichtsstand: Hamburg

Vorstandsmitglieder:
Prof. Dr. Jvrg F. Debatin (Vorsitzender) Dr. Alexander Kirstein Ricarda
Klein Prof. Dr. Dr. Uwe Koch-Gromus




-- 
Pflichtangaben gemäß Gesetz über elektronische Handelsregister und Genossenschaftsregister sowie das Unternehmensregister (EHUG):

Universitätsklinikum Hamburg-Eppendorf
Körperschaft des öffentlichen Rechts
Gerichtsstand: Hamburg

Vorstandsmitglieder:
Prof. Dr. Jörg F. Debatin (Vorsitzender)
Dr. Alexander Kirstein
Ricarda Klein
Prof. Dr. Dr. Uwe Koch-Gromus