Overall workflow for the Chloroplast 2010 project pipeline
More detailed protocols for the individual assays can be found at the project website.

1. Newly available homozygous mutant lines at ABRC are searched against the project’s list of target genes and stocks annotated as mutant in any of our genes are ordered.
   
   
2. New stocks are logged into our database, 3-4 seeds sown in each of two pots on soil-less mix, stratified at 4°C and then grown under 120 μmol photosynthetically active radiation (cool white with supplementary incandescent) in BioChambers Arabidopsis chambers at 21°C and 16 hour photoperiod. Each pot is thinned to one plant and grown for seed in Aracons. The sibling plants are grown in separate flats throughout their life cycle. On day 21 after germination for plants grown at 16 hr photoperiod, and day 28 after germination for plants grown at 12 hr photoperiod, plants are individually photographed and controlled vocabulary descriptions noted in the database.
   
   
3. The plants are genotyped to test for homozygosity and lines with no homozygous mutant plants are discarded. Results are recorded in the database. Note that approximately 25% of the mutants that we obtain are not true breeding for the allele based upon our genotyping results.
   
   
4. Seeds from both siblings are harvested, and used directly for seed phenotypic assays (amino acids, carbon / nitrogen, morphology, starch). Seeds are also sown into pots (one pot per sibling for a total of two pots per mutant) and stratified at 4°C. The plants are grown under 12 hour photoperiod under 120 micromol photosynthetically active radiation (cool white with supplementary incandescent) in BioChambers Arabidopsis at 21°C. Siblings are grown in different flats, and sown on successive days so that their growth is staggered by one day.
   
   
5. Images and description data are collected for plants grown at 12 hr photoperiod on day 28. The next day, leaves are harvested for morning starch, amino acids, fatty acids and afternoon starch, and three days later for chloroplast morphology. We strive to run each assay at the same time of the day to decrease impacts of diurnal regulation on the data.
   
   
6. If one of the sibling plants does not grow or one or more phenotypic assays yields unsatisfactory results, one or two plants are grown again to ensure that there are duplicate data points for all assays. Mutants whose siblings give conflicting results are generally retested.
   
   
7. Occasionally, mutants with interesting phenotypes are followed up by ordering non-homozygous lines of second or third alleles, and homozygotes are identified by genotyping.
   
   
8. Measurement of chlorophyll fluorescence is performed on plants grown in 96-well flats especially for this assay.
   
   
9. Routine genotyping of all putative insertion mutant line plants grown for seed (see step 3, above) began two years into the project. Prior to that single leaves of the plants used for genotyping were archived on Whatman FTA cards, and these samples were subsequently genotyped.

Accessing the Chloroplast 2010 Database: Simple Query
The page used for entering the 2010 Analysis Database − it is used to find information for lines associated with a specific gene.

• Use Search by Query Term(s) to look for information about one or a group of specific SALK mutant lines, Arabidopsis genes (AGI numbers or TAIR 8 names), pots or flats of plants.
  
  
• To return results, click Submit. The database will return a table with all instances where a plant was sown or seed was harvested for mutants in the gene(s). This allows the user to explore all of the data for mutants of a given locus. Note that if none of the lines sown were found to be homozygous, phenotypic data are unlikely to be available.

Accessing the Chloroplast 2010 Database: Advanced Query
This page is used for entering the 2010 Analysis Database - it is used for finding mutants with specific phenotypes or combinations of phenotypes.

• Use the Search by Assay Results feature to search for genes with mutant plants associated with specific phenotypic abnormalities, or combinations of changes. The Chloroplast 2010 Database has three general types of phenotypic data: photographs (not searchable), controlled vocabulary descriptions and quantitative metabolite data. Note that the 'AND' and 'OR' Boolean logic at the top applies to all assays and filter characteristics selected below.

• Phenotypic queries: To find examples of mutants associated with an altered phenotype, click Abnormal. This often activates other search options, and at least one of these characteristics must be clicked (for instance, if you ask for Abnormalities in Whole Plant Characteristics, several different controlled vocabulary attributes will become available, and at least one of these must be selected for the data to be filtered further). Data for controlled vocabulary attributes were entered by the technician who ran the assay.
  
  
• Searching quantitative data − For quantitative metabolite data (Amino Acids, Fatty Acids and Carbon/Nitrogen), choosing Abnormal will reveal a set of statistical cutoff options. For example, choosing Abnormal in Carbon / Nitrogen Characteristics will require that you select a Minimum z-Score cutoff.
  
  
  • What do these z-Scores mean? − As previously reported by Jander et al (2004; Plant Journal 39:465-475), we have found that detection of mutants is aided by analysis of a cohort group of plants that were grown together. In the Chloroplast 2010 project we grow plants in flats of up to 32 individuals, and calculate a flat median for each trait. Median was chosen rather than mean to minimize the effects of extreme outliers that are sometimes detected. Deviation from the median was calculated (Median Absolute Deviation or MAD) and reported as z-scores. A z-score of 2, for example, is equivalent to 2 deviations from the flat median. For more information see Lu et al. (2008) Plant Physiol. 146:1482-1500.
    
    
  • For example: choosing a Minimum z-Score of 1 for Carbon / Nitrogen and hitting Submit in the lower right corner of the screen will return a list of genes, mutant alleles, summary annotation for the gene (Definition) and information about how many samples were run through the various assays. If an unexpectedly large or small number of genes are returned, click back the arrow on your browser and refine your search.
    
    
  • Choosing Leaf of Seed Amino Acid Assay or Leaf Fatty Acid Assay will give choices of metabolites and Minimum z Score. Remember that OR/AND Boolean logic (query conditional at the top of the query area) will apply to any trait that you choose.
    
    
  • Selection of 'All Normal' in the fields that query quantitative data (Leaf or Seed Amino Acid, Seed C/N ratio, Leaf Fatty Acid) will filter for samples having < the default threshold z-score value for that assay. To see the current threshold setting, click the 'Abnormal' radio button for that assay.

Accessing the Chloroplast 2010 Database: Advanced Query Results
This table shows genes with mutants that have one or more sample that was found to be abnormal for the phenotype or combination of phenotypes used in the query. It is important to remember that insertion mutant lines often have insertions and point mutations in other genes. That is a main reason why we strive to run multiple independent mutants for each gene on our target list, as they become available.

• This page summarizes the results of a query, with information about the number of alleles for which one or more plants had a phenotype that matched your query. You can navigate multiple pages at the bottom left of the page (Previous, Next and page numbers). Clicking on the table headings will sort the table based on that column's contents (for numbers from high to low or low to high, or alphabetically). Clicking on the AGI (i.e. At) number will open a new page with detailed information for this gene. The column on the far right represents a TAIR summary annotation for that gene.
  
  
• It is important to note what this table does and does not tell you about the data in the Chloroplast 2010 database.
  
  
  • What the presence of a gene in this table tells you: At least one sample associated with a mutant of this gene was abnormal based upon your search criteria.
    
    
  • What it does not tell you: The presence of a gene locus on this table does not mean that all samples that were assayed were abnormal based upon your search criteria. A single outlier sample out of many run is enough to cause the gene to appear on the table.
    
    Similarly, it does not mean that all mutants listed in 'Source Designations' had one or more samples that are outliers. If two alleles are listed, it is possible that only one of the mutants had the mutant phenotype.
    
    
  • Why do we present the data in this way? We choose to have relaxed criteria for presenting the results of trait-based searches for a number of reasons.
    
    
    1. Many of our assays are quantitative and there is no one absolute criterion for deciding whether an individual is 'abnormal'. Different assays have different amounts of environmental variability, and we would prefer to offer you the chance to look at a larger dataset and make your choices about which genes are interesting to you.
       
       
    2. In cases where there are multiple mutant alleles for a given gene, differences in the t-DNA insertion sites will influence the severity of the phenotype. If you are interested in a gene, we urge you to look up the position of the sequenced alleles at the t-DNA Express website. It is also crucial for you to obtain samples of the mutant seed from ABRC or NASC and to test the genotyping and phenotypes of those seeds. Sorry - we do not have resources to distribute seed.
       
       
    3. What do the numbers in the table mean? This is an attempt to answer the question: 'How strong is the phenotype for the mutants in this gene?'. High numbers are consistent with multiple samples of one mutant satisfying your search criteria. Having mutants with multiple mutant alleles will also give a higher score. This system is still a work in progress, and we know it is far from perfect. If you have an interest in this sort of analysis let us know - we would like to get your advice.
       

  We are experimenting with different ways to evaluate the magnitude and consistency of phenotypes. Please note that these features will change over time.

Accessing the Chloroplast 2010 Database: Gene Page
This allows you to review data related to plants associated with mutations in specific genes.

• This set of tools allows visualization of a variety of data about single genes and mutants annotated as defective in the gene. It is designed to allow the user to choose which data to view, and the combinations of data types. Clicking on a tab (Whole Plant Morphology, Chloroplast Morphology, etc.) changes the information that is presented for that gene. Tabs with white text and darker green background are active. Clicking on a pot number allows you to explore data for the individual plant or seed pool.
  
  
• Top left panel: This area is designed to show photographs and graphs. The display of data is controlled by clicking on a tab at the top of the window.
  
  
• Top right panel: This area will be used for different information as time goes on. At the time of writing, it is only used in the Chlorophyll Fluoresence Image view, which is activated by clicking on the Fluorescence Sample Info tab.
  
  
• Middle panel: This area is for data summary and is used to view controlled vocabulary phenotypic descriptions (Whole Plant Morphology, Chloroplast Morphology, Seed Morphology, Starch and Fluorescence Data Summary, for examples) and simple quantitative data (for instance Carbon / Nitrogen Ratio with z-score). The data display is controlled by clicking on a tab above the window.
  
  
• Bottom panel: Along the bottom of the window is displayed a summary table describing the samples associated with this gene.
  
  
  • Pot: Refers to the plant assayed or the plant that was harvested for the seed assay. Clicking on the 'P number' causes data for that sample to be displayed in the two panels above (which data are displayed depends upon the tabs that are active).
    
    
  • Planted: Information about when the seeds were sown.
    
    
  • Genotyping Data: Describes whether the plant or line was confirmed as homozygous for the insertion allele. N/A = no sample is available to genotype; Not Done = no assay done on this sample.
    
    
  • Allele/Source: The name of the allele that was tested. More information about insertion alleles should be available by clicking on the hotlink, which will cause you to navigate away from our site.
    
    
  • At a Glance: Summary of the phenotypes that deviate from 'normal'. Mousing over the boxes causes a dialog box listing the phenotype name to appear above this column. We are experimenting with different ways to evaluate the magnitude and consistency of phenotypes. Please note that these summaries should be interpreted cautiously and the parameters and approaches will change over time. Traits with multiple variables (for example, the twenty measurements each of leaf and seed amino acids) will be overrepresented in these summary statistics.
    
    
• Leaf Tissue, Seed Tissue and Fluorescence Sample Tabs: Just above the summary table. These tabs have an important function: they control the types of data that are displayed in all three windows. When Plant Data is active, the top fields emphasize vegetative leaf data, whereas, when Seed Data is active, the other fields emphasize seed morphology and chemistry. Clicking the Fluorescence Sample Info tab reveals data from this specialized assay. These are false color images for plants in a set (top left), controlled vocabulary descriptions of the results of each assay (middle) and summary information about the samples assayed for this mutant (bottom panel). False color images for each of the assays (Fv/Fm before high light, right after high light and several days after high light as well as NPQ before high light) can be selected by clicking on the hotlinks above the image panel. When the false color images are displayed, the window to the right of the image will show information about the plants in each frame and the gene annotated as being defective in the plant.

• External Hotlinks: Top right of the window.
  • TAIR opens a tab displaying the TAIR Gene Search Page.
  • BAR eFP opens the BAR eFP Browser for genes that are represented on the Affymetrix ATH1 GeneChip. This provides an overview of mRNA expression data.
  • Correlated mRNA Analysis opens the BAR Expression Angler tool for genes represented on the ATH1 GeneChip. This provides an indication of genes with similarly regulated mRNAs.
    
    
• General notes about the assays: Nearly all of the assays are performed on plants, or seeds from plants grown in sets (flats) of 32. To control for differences caused by microenvironment, or subtle changes in growth conditions over time, both qualitative and quantitative assessments of phenotypes are based upon comparisons with the other plants in that flat. For instance, z-scores are calculated based upon statistics for that flat; assessments of heterogeneity of seed morphology is in comparison to other seed pools from that flat, etc.
  
  Information about the protocols can be found by clicking here.
  
  
• Details of the tabs that run across the top of the top right insert:
  
  
  • Whole Plant Morphology: Image and controlled vocabulary are for a 21-day old plant (16/8 photoperiod) or 28-day old plant (12/12 photoperiod). Rosette size is a semi-quantitative description of the length of the largest leaves of the rosette:
    
    
• Normal: 4.0-6.0 cm diameter plant
• 1: 2.5-4.0 cm diameter plant
• 2: 1.25-2.5 cm diameter plant
• 3: <1.25 cm diameter plant


• Chloroplast Morphology: Shows fixed mesophyll cell photographs and controlled vocabulary descriptions for two or three different tissue types.
  
  
• Seed Morphology: Image and controlled vocabulary are for mature dried seed from plants grown under 16h photoperiod. A 0.5mm bar is shown for reference − this is the length of a typical Col-0 wild-type seed from our collection.
  
  
• Seed C/N: (ratio of carbon to nitrogen) is presented as a surrogate for direct measurement of storage oil and protein. Increased ratios could indicate accumulation of excess oil or reduced protein and visa versa.
  
  
• Leaf Starch and Seed Starch: Controlled vocabulary descriptions of the results of iodine staining for starch.
  
  
• Leaf Amino Acid and Seed Amino Acid: A graph documenting the z-score of each of the metabolites measured. Mousing over a bar reveals the calculated z-score compared based upon the flat median. A red bar indicates that the z-score is outside the threshold of the graph. Clicking Play (>) cycles the data in the graph through every analyzed pot in the bottom panel so that discrepancies and patterns are easy to discover. The Y scale and speed of cycling (Delay) can be controlled.
  
  
• Leaf Fatty Acid: A graph documenting the z-score of each of the metabolites measured. Placing the cursor over a bar reveals the calculated z-score compared with flat median and the number of prcocess replicates for that measure (typically two independent samples per plant). A red bar indicates that the z-score is outside the threshold of the graph. Clicking Play cycles the data in the graph through every analyzed pot in the bottom panel so that discrepancies and patterns are easy to discover. The Y scale and speed of cycling (Delay) can be controlled.
  
  
• Leaf Fatty Acid Images: A reconstruction of the gas chromatography-flame ionization detection (GC-FID) trace. These are included to provide an opportunity to search for novel peaks and evaluate the quality of the GC run.