Skip to end of metadata
Go to start of metadata

In this tutorial, we will upload the data from supplemental table 1 from the paper by Grice et al. (2007) Transcriptional profiling of C57 and DBA strains of mice in the absence and presence of morphine.   BMC Genomics. Mar 16;8:76.

  1. Download the supplemental data from the paper.  Note that you can use either the journal copy or the Pub Med Central copy to access the data.  We usually prefer the Pub Med Central, because it is ensured to be open access for all users and sometimes makes it easier to provide direct links to material within the paper rather than a general link to the paper itself.  However, in this case, it turns out that the only way to directly link to the supplemental data table is through the BMC Genomics link, rather than Pub Med Central.  We have also provided this sample data set on the 3rd worksheet of the DRG Data TemplateDDG Data Template - External User.xls.  Also note that if the data are in pdf format, you will have to extract the data into a spreadsheet format before beginning the process.  See PDF converter tool instructions .
  2. Review the data:  Note that in this table, there are actually 3 different sets of data reported, which may be viewed by scrolling through the entire spreadsheet:  A:  Genes showing differential expression between vehicle-treated C57 and DBA strains; B:  Genes showing differential expression between morphine-treated C57 and DBA strains;  C:  Genes showing differential expression between vehicle and morphine-treated DBA mice

The Excel spreadsheet should look like this: 


Genes showing differential expression between vehicle-treated C57 and DBA strains




Probe set

Gene Title

        C57 mean

      C57 SE

     DBA Mean


  Fold   change

     P value


phosphatidylinositol transfer protein, beta








riken cdna e130013n09 gene








actin related protein 2/3 complex, subunit 5








aurora kinase a interacting protein 1








peroxiredoxin 2








tbc1 domain family, member 15








tumor necrosis factor alpha induced protein 6








riken cdna 2810433d01 gene







  2. Fill out article information in template: Note that many supplemental data do not contain an explicit reference to the article from which they originate. So in the DRG, we must specifically add the link to the article.  When completed, it should look like this:


    Field Name

    Your Values Here





    Article Title:

    Transcriptional profiling of C57 and DBA strains of mice in the absence and presence of morphine



    Grice DE, Reenilä I, Männistö PT, Brooks AI, Smith GG, Golden GT, Buxbaum JD, Berrettini WH.


    Abbrev Pub:

    Grice et al., BMC Genomics. 2007


    GEO Accession #:



    Article URL:

Note that in this case, the data were not deposited in Geo according to what is stated in the article, so this field is left blank.

4.  Insert Table descriptive information:  In this step, you will provide the overall description of the data that are contained in Supplemental Table 1.  However, because the authors provided 3 different conditions inside of S1, we actually have to turn this into 3 different tables.  So, S1 table A is entitled:  Genes showing differential expression between vehicle-treated C57 and DBA strains;  S1 table B would be entitled:  Genes showing differential expression between morphine-treated C57 and DBA strains etc.

Here is the metadata for S1 table A:


Field Name

Your Values Here


Table Number:



Table title:

Genes showing differential expression between vehicle-treated C57 and DBA strains.


Table Subtitle:


Table URL:



Gene expression


Associated Images / Bar graphs:


Images / Bar graphs URL:


Methods URL:


*Assay / Protocol Type

DNA microarray


Platform (Company and product):

Affymetrix MG_U74A high-density oligonucleotide array


*Organism Species:



Experiment vs. Control:

DBA/2J vs. C57BL/6

For all of the fields designated by an *, the values are taken from the controlled vocabularies provided by NIF.  A listing of the most popular terms may be found here .

5.  Start transferring data values to the DRG template:  The DRG template provides column headers for the data and metadata required for their interpretation for the types of experiments that we currently support (Refer to Assay/Protocol Type). In this case, it's a DNA microarray experiment.  The first column of the DRG template should be left blank;  it just contains a description of what the column headers mean.  Save the file using using the author name and table:  DRG Data Template - Grice_TableS1.xls

Section 1:  Gene information:  In our example table, the paper provides values for the gene name and the probe ID.  In this example, Gene name is called "Gene title" and Probe ID is called "Probe set".  Select the values associated with section A of the table and copy and paste them into the template.  If you leave a space at the top, make sure you do that for all of the columns.  The Gene Symbol and the Gene ID were not provided, so leave these blank.

Section 2:  Anatomical structure:  In this case, reading the results and methods section provides the information that the results in Table S1 all came from the nucleus accumbens.  The preferred labels for anatomical structures referenced in papers should be drawn from the Neurolex.  Enter the value:  Nucleus accumbens for all of the results in Table S1A.

Section 3:  Organism information:  The methods section reports that:  "Mice were obtained from the Jackson Laboratory. Animals were male, 8--12 weeks old, with a weight range of 18 to 28 grams."  Enter the value "male" under gender for all rows.  Enter 8-12 weeks for the age for all rows.  Add 18-28 grams for all rows.  We do not standardize on units at this time, so enter the values as stated.   Look up the age ranges for the NIF annotation standards on age classification.  NIF defines an adult mouse as > 50 day post natal so mice 8-12 weeks of age would be classified as Adult so use the label "Adult mouse".

Section 4:  Experimental Organism Method Information:  According to the abstract,  "The mouse C57BL/6 (C57) and DBA/2J (DBA) inbred strains differ substantially in many aspects of their response to drugs of abuse".  Even though the methods do not provide the exact strain name, they do state that they purchased the mice from Jackson Laboratory and do state the strain in the abstract.  Use the preferred label from the Neurolex for strain names, as there are many ways of saying the same thing.  Enter DBA/2J into the experimental column into every row.  We consider it experimental because the fold change is presented as DBA/2J over C57BL/6, even though the order of the columns are reversed in the spreadsheet.  We leave the experimental strain characteristics blank in this case, because it wasn't a mutant strain, but you can include notes as necessary.  Enter C57BL/6 as the control strain into every row.

Section 5:  Experimental Drug Method Information:  Enter the treatment given to the experimental and the control groups.  In this case, reading the methods shows that "Vehicle" treatment in the title of the results set refers to a placebo pellet:  "Mice were obtained from the Jackson Laboratory. Animals were male, 8--12 weeks old, with a weight range of 18 to 28 grams. Animals underwent surgery for subcutaneous implantation of either a control or morphine pellet (either the "Placebo to Morphine Base Implant Pellet" or the 25 mg "Morphine Base Implant Pellet," both the kind gift of the National Institute on Drug Abuse) on the dorsal aspect of the neck."  Enter "Placebo pellet" under both Experimental drug 1 and Control drug 1 and copy it to all rows because there was no difference in treatment between the groups.

Section 6:  Experimental Results: Paste the values for the DBA/2J into the experimental values.  Because the data are presented as mean +/- standard error, you must do an extra step to concatenate these values together.  The easiest way is to start a new Excel spreadsheet, and paste the mean values into column A and the SE values into column C.  In column B, type a space followed by an open parentheses "(".  In column D, put a closed parentheses ")".  Copy both of these values for all rows.  Then, in column E, use the =concatenate formula to put them all together:
The results should look like this:
Copy the concatenated values, and using the "Paste special > values" function, transfer to the DRG template.  For the exp/cont ration, paste the ratio values expressed in the "Fold change" column of  the original data.  Add the text modifer "Fold change" to the Exp/Contr modifier column, to indicate the meaning of the ratio.  For the p values, paste in the p value column.  The statistical test column provides a description of the statistical analysis used.  In this case, you can put in as much detail as you think appropriate:  e.g., two-way analysis of variance (ANOVA) using dChip, as the analyses are usually described in detail in the paper

Section 7:  Conclusion:  In this column, we place the qualitative annotation standards gene expression to capture the results of the gene expression analysis.  In this table, the paper states that the genes listed are differentially expressed, i.e., the fold change between groups reached statistical significance.  The table doesn't explicitly state whether expression is increased or decreased, but since they provide the values, direction of change, i.e., either positive or negative, and the p value, we can map their results to our qualitative expression level categories. For each negative value, we will put in "Decreased expression" and for each positive value, we will put in "Increased expression."  Again, please consult with the NIF annotation standards for analysis results and put in the exact phrase used. 

Section 8:  Other.  This field can be used to store any comments or questions about the data.  In this paper, the curator noted that two values were duplicated for genes expressed sequence al033311 and cathepsin e.  It seems unlikely that the same exact values would be found for the genes in both strains.  The curator contacted the authors, but they no longer have access to the data.  So, we can put in a comment here that it looks like there might be an error in the data.

You can see what a copy of the data look like in the 2nd tab on the worksheet (Example DRG Template).  Once the data conversion is complete, upload the template to the Wiki or send an email to  The data will undergo a review by curators prior to upload.  The curation includes both a review of the data to ensure consistency, and a mapping of the content to the NIF ontology identifiers.  Once this is complete, the data will be made available for query through the NIF.



  • No labels