Unmatched MBF data value while analyzing SNP markers

From: Fikret Isik <fisik_at_UNITY.NCSU.EDU>
Date: Sun, 6 Dec 2009 21:29:50 -0500

Dear ASReml users,

I am analyzing SNP markers to examine their association with phenotype
(growth of trees). There are 136 clones. Each clone has multiple copies
(genetically identical) tested across three environments (LOC).

Both phenotypic and marker data have missing phenotypes and
markers for some clones.

The .as job file and the mixed model is:

Title: CFor clones
#101,9,11,J,Holly Hill-01,2,35.9,7,0,3.4838
#101,9,11,J,Holly Hill-01,4,38.3,6.6,0,3.3058
#101,9,11,J,Holly Hill-01,6,27.7,5.85,0,1.8929
  clone !P
  p1 !I
  p2 !I
  famid !A
  LOC !A 3
  REP *
CF_Pedigree.txt !SKIP 1 # Clones are included in pedigree file
!CYCLE 1:449
!MBF mbf(clone,1) CF_marker1.csv !SKIP 1 !RFIELD $I !RENAME SNP$I !DDF 2
!FCON # marker data

HT_08 VOL ~ Trait Tr.LOC Tr.SNP$I ,
             !r Trait.clone Trait.ide(clone) Trait.REP.LOC
3 2 3
780 0 IDEN
Trait 0 US 11.5 0 3.17 !GU # HollyHill
898 0 IDEN
Trait 0 US 14.1 0 2.79 !GU # Oliver_Orchard
393 0 IDEN
Trait 0 US 11.0 0 2.79 !GU # Oliver_RR

Trait.clone 2
Trait 0 CORGH !+6 !GU #or !GU
clone 0 AINV

Trait.ide(clone) 2
Trait 0 IDEN 0.2 0.1 0.3
ide(clone) 0 IDEN

Trait.REP.LOC 2
Trait 0 IDEN 0.2 0.1 0.3

The model runs fine but the .asr output file reports the following
message for hundreds of data points for each marker:

Unmatched MBF data value 100.00 in data record 736
            1 101.00
Unmatched MBF data value 100.00 in data record 737
            1 101.00
Unmatched MBF data value 100.00 in data record 738
            1 101.00

I checked the ASReml3 manual (the latest version) but could not find any

Q1) Is this because of missing phenotype records fro some markers?
Q2) How stringent the program is about the missing values in the
phenotypic and marker data?
Q3) What would be solutions?

I appreciate any feedback.

Some of you already know this: ASReml did not accept '.' or 'NA' as
missing values in the marker data file. However, when I converted missing
values to BLANK it worked.

Fikret Isik
Received on Mon Dec 06 2009 - 21:29:50 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)