3. In the case of other population such as RILs, BILs, and GWAs,
less replication is needed than in the ILs since in these populations genetic variance is represented in multiple lines, as
opposed to a single line, within the population [28].
4. In order to minimize the variation introduced by sampling,
there are several crucial points to take into consideration during
harvest (see Note 6).
3.3 Sample
Processing and
Extraction
1. Harvest plant organs (e.g., leaves, flowers, or fruits) and freeze
them immediately in liquid nitrogen to be stored at –80
C, or
grind them immediately to a powder and start the extraction
(see Note 7).
2. Before extraction quality control (QC, see Note 8) samples
should prepared by pooling aliquots of individual study
samples.
3. Distribute the QC samples across all machine-batches.
4. Extract, derivatize, and analyze aliquots thereof at the exact
same time as the individual study samples [28].
3.4 Sample
Preparation and
Analysis
1. In LC-MS extract directly introduced to the apparatus (see
[29]).
2. In GC-MS based metabolomics, derivatization is needed which
involves several general steps ([17, 18, 30, 31] for details).
3. We recommend to divide the samples in batches so that each
batch contains 50–80 samples with ample QC samples
distributed across the sequence run [28].
3.5 Data Processing
1. Once samples are analyzed, automatic data processing tools
are required for peak picking and mass peak alignment (see
Note 9).
3.6 Data
Normalization
1. Data normalization is a crucial step in any metabolomics study.
First step is normalizing using an internal standard; this reduces
the variation in sample extraction, and compared for machine
drifts.
2. Further normalization to sample weight and QC samples is
additionally useful (see Note 10).
3.7 GWAS Mapping
The basic principle of genome-wide association studies (GWAS),
which were initially developed for use in medical genetics, is that
the incidence of nucleotide polymorphisms is associated with the
presence of variance, overcoming the limitations of using ILs and
RILs (see Note 11). The step-by-step instructions for GWAS
mapping are as follows:
398
Si Wu et al.
less replication is needed than in the ILs since in these populations genetic variance is represented in multiple lines, as
opposed to a single line, within the population [28].
4. In order to minimize the variation introduced by sampling,
there are several crucial points to take into consideration during
harvest (see Note 6).
3.3 Sample
Processing and
Extraction
1. Harvest plant organs (e.g., leaves, flowers, or fruits) and freeze
them immediately in liquid nitrogen to be stored at –80
C, or
grind them immediately to a powder and start the extraction
(see Note 7).
2. Before extraction quality control (QC, see Note 8) samples
should prepared by pooling aliquots of individual study
samples.
3. Distribute the QC samples across all machine-batches.
4. Extract, derivatize, and analyze aliquots thereof at the exact
same time as the individual study samples [28].
3.4 Sample
Preparation and
Analysis
1. In LC-MS extract directly introduced to the apparatus (see
[29]).
2. In GC-MS based metabolomics, derivatization is needed which
involves several general steps ([17, 18, 30, 31] for details).
3. We recommend to divide the samples in batches so that each
batch contains 50–80 samples with ample QC samples
distributed across the sequence run [28].
3.5 Data Processing
1. Once samples are analyzed, automatic data processing tools
are required for peak picking and mass peak alignment (see
Note 9).
3.6 Data
Normalization
1. Data normalization is a crucial step in any metabolomics study.
First step is normalizing using an internal standard; this reduces
the variation in sample extraction, and compared for machine
drifts.
2. Further normalization to sample weight and QC samples is
additionally useful (see Note 10).
3.7 GWAS Mapping
The basic principle of genome-wide association studies (GWAS),
which were initially developed for use in medical genetics, is that
the incidence of nucleotide polymorphisms is associated with the
presence of variance, overcoming the limitations of using ILs and
RILs (see Note 11). The step-by-step instructions for GWAS
mapping are as follows:
398
Si Wu et al.