Using the Gene Expression Data Query Form

This help document answers the following questions:

• What is the purpose of the Gene Expression Database?
• What can I use the Gene Expression Data Standard Search to find?
  • What values are acceptable in each of the Standard Search fields?
• What is the purpose of the Differential Expression Search?
  • What values are acceptable in each of the Differential Search fields?
• What is the purpose of the Expression Profile Search?
  • What values are acceptable in each of the Profile Search fields?
• Can I perform batch searches for expression data with multiple genes?
• Can I edit my original options and requery?
• What are the advantages of the different results tabs?
• How do I interpret the query results summary page?
  • What do the colors in a Matrix mean?
  • How are expression results propagated through the matrix anatomy hierarchy?
• How do I filter my results?
• Can I get my results as a tab-delimited text file?
• Can I create a heat map for my RNA-Seq results?
• Can I send the results of my search to the MGI Batch Query or MouseMine?
• Why is there a link to Gene Expression Literature records in my query summary?
• Which expression data query parameters generate an expression literature search? Which do not?
• How often are annotations added?
• Are there examples of query results?

What is the purpose of the Gene Expression Database?

The Gene Expression Database (GXD) is designed to integrate many different types of endogenous gene expression data from the mouse in formats appropriate for comprehensive analysis. Query results, such as the developmental stage and tissue of expression (or non-expression), the genetic origin of the sample, and the numbers and sizes of detected bands, are described together with the molecular probe, the expression assay type, and the experimental conditions used. Expression patterns are described using an extensive, hierarchical dictionary of standardized anatomical terms, making it possible to record expression results from assays with differing spatial resolution in a consistent and integrated manner and to analyze expression patterns at differing levels of detail. Whenever possible, text annotations are complemented by digitized images of the original expression data.

To learn more about GXD and how we acquire data, please see About GXD.

See also the MGI Glossary for unfamiliar terms and acronyms.

What can I use the Gene Expression Data Standard Search to find?

The Gene Expression Data Query form Standard Search tab allows you to ask questions that can be basic or complex. Examples of the type of questions you can ask are:

• When and where is En1 expressed?
• What genes are expressed in the heart at E8.5 in wild-type mice only?
• Is Htr7 expressed in the brain?
• What members of the Wnt gene family are expressed in TS 16?
• When and where are genes involved in axon guidance expressed?
• What genes have been shown NOT to be expressed in either the diencephalon or any structure within the diencephalon?
• What genes involved in cell adhesion are expressed in stomach mesenchyme?

(See Are there examples of query results? for the form field values for these queries.)

If you are interested in doing queries that ask what genes are expressed in some anatomical structures and/or developmental stages but not in others, you should use the Differential Expression Search tab on this form.

Top

What values are acceptable in each of the Standard Search fields?

Genes | Genome location | Anatomical structure or stage | Mutant / wild type | Assay types

Genes	This section offers a choice of two fields. Nomenclature search. Use the left hand field to search for expression data for a mouse gene or set of genes with a similar nomenclature. The gene nomenclature field supports an asterisk as a wild card. A search for pax* finds expression data for gene symbols and names that begin with pax. A search for *box* looks for nomenclature that contains box. This search finds the arginine decarboxylase gene, homeobox genes and paired box genes. You can also search with a comma-separated list of genes. Use the Batch Search tab for a column of genes or IDs. See Using the Gene Expression Batch Search for details. A search for growth factor finds nomenclature containing growth AND factor. Synonyms (unofficial gene names used in publications and former official gene symbols) are also searched. This field searches only alphanumeric characters. Single and double quotes, commas and hyphens are ignored. Vocabulary term search. Use the right hand field to find expression data for genes defined by their function, phenotype or disease model association. Begin typing a term of interest and an autocomplete list activates on the second character. The list shows all the vocabulary terms and their synonyms that match your search characters (shown in bold letters) and the number of genes annotated to each term. Terms are greyed out if there are no genes annotated to the term or if there are no expression data associated with the genes. You can type the beginnings of multiple terms to narrow the list (example: axon guid). You must select a term. There is no free text searching in this field. You can only select one term, though choosing a term includes results for its subterms. For example, selecting decreased bone mass also includes results for its subterms, decreased compact bone mass, decreased trabecular bone mass and osteoporosis. The autocomplete field ignores non-alphanumeric characters such as " and -. If you are unsure of the proper terms to use, use the appropriate browser to find vocabulary terms and their definitions: Function: GO Browser Phenotype: Mammalian Phenotype Browser Disease: Disease Ontology Browser Note that you may not submit a search with values in the Nomenclature search and the Vocabulary Term search at the same time.
Genome location	You can enter a list of mouse genomic regions (e.g. 2:105668896-108698410, Chr12:28553755..28867491). You can enter multiple regions separated by commas, spaces and/or returns. You can also search with a whole Chromosome (e.g. 19 or 19: or chr19). Use MT for the Mitochondrial Chromosome. Toggle the default, bp (base pairs), to Mbp, to search by mega base pairs. (e.g. 12:28.55-28.89)
Anatomical structure or stage	Click to limit your search to assay results where expression was detected in, not detected in, or either. The default value is either. You can choose to use one or both fields in this section. Anatomical Structures The anatomical structures are taken from the Mouse Developmental Anatomy Ontology. Begin typing an anatomy term of interest and an autocomplete list activates on the second character. The list shows all the anatomy terms and their synonyms that match your search characters. The term is followed by the Theiler Stages for which the structure exists (e.g. lung lobe TS20-28). You must choose a term from the autocomplete list (i.e., there is no free text entry in this field). The ordering on the drop down is as follows: exact match to structure exact match to a synonym "begins" match to a structure "begins" match to a synonym and then other matches For each developmental stage, the Anatomical Dictionary is organized as a hierarchy of structures. By default, the search includes substructures. This means if you search for gene expression in, for example, brain, your search will also return annotations for all substructures of brain, such as rhombencephalon, tectum and others. If you are unsure of the most appropriate structural term to use, browse the Mouse Developmental Anatomy Ontology. Developmental Stages Use this field to select one or more Theiler stages (TS) to focus your search on a particular stage of embryonic development. Alternately, you can choose the Use Ages tab and limit your search to the general terms, Embryonic or Postnatal, or choose specific days post conception (dpc). Note: The Theiler system organizes development into stages defined by the appearance of specific developmental features. Embryos of the same gestational age can vary considerably with respect to development. Consequently, a Theiler stage does not precisely correspond to a particular age. For example, TS 21 applies to embryos between 12.5 and 14.0 days post conception (dpc), while TS 22 applies to embryos between 13.5 and 15.0 dpc. To select multiple stages or ages: On a Macintosh, hold the command (Apple, ⌘) key down and click on the desired stages; OR hold the shift key down and click on two stages to select those and all stages in between. On a PC, hold the control key down and click on the desired stages; OR hold the shift key down and click on two stages to select those and all stages in between. If you mouse-over a stage, a pop-up will describe a few defining features of that stage, such as "first somites, late head fold" for Stage 12. You can browse Stage descriptions for a list of the defining features for each Theiler stage.
Mutant / wild type	Select Specimens mutated in gene to search for expression data obtained from mutant specimens. Use this field to enter the symbol or name of the mutated genetic marker. The system searches for current symbols/names and synonyms of the mutated gene. Use this search field as you would the Genes' nomenclature search at the top of the search form. Select Wild type specimens only if you wish to exclude mutant alleles from your search. This search will still include knock-in reporter genes used in in situ reporter assays. Homozygous mutant specimens are not included in this search. Select All specimens, the default, to include all mutant and wild type specimens in your search.
Assay types	Check list of assay types. Use this section to limit your search to assays of one or more selected types. The default is to exclude RNA-Seq assays from the search. Use the in situ, blot and whole genome assay check boxes to quickly select or deselect types of assays.

Top

What is the purpose of the Differential Expression Search?

The Differential Expression Search is specifically designed to allow you to search for genes expressed in some anatomical structures or developmental stages but not in others. Examples of the questions you can ask are:

• Which genes are expressed in the forebrain and not expressed in the midbrain?
• Which genes are expressed exclusively in the eye and its substructures?
• Which genes are expressed only at Theiler Stage 15?

RNA-Seq data are not yet included in these searches.

Top

What values are acceptable in each of the Differential Search fields?

Anatomical Structure | Developmental Stage

The Differential Expression form lets you query for genes expressed in some anatomical structures and/or developmental stages but not in others.

• To specify structure(s), type an entry in the field or make stage selections from the list. You may use both fields simultaneously.
• Structures and/or stages entered in the upper section query the database for assay results demonstrating the expression of genes in those structures or stages.
• Structures or stages entered in the lower section query for the absence of expression of genes in those structures or stages. Note: If the structure or stage in the lower section has conflicting expression results annotated to it, i.e., results showing both the presence and absence of expression of a particular gene, that gene is considered to be expressed and is not returned by this query. Expression results that are either Ambiguous or Not Specified are ignored by this query, i.e. they are treated as neither present nor absent.
• These searches will return a list of genes whose expression was shown to be absent (not detected), as well as genes whose expression has not been analyzed or recorded in the database for the specified structures and stages.
• The search is limited to results from wild-type specimens. For the purposes of these queries, results from in situ reporter (knock-in) experiments will be considered wild type if the results are derived from a heterozygote.

Field	Description
Anatomical Structure	The anatomical structures are taken from the Mouse Developmental Anatomy Browser. Begin typing a term of interest and an autocomplete list activates on the second character. For each stage, the Anatomical Dictionary is organized as a hierarchy of structures. Thus the Find genes where expression is detected in, search includes substructures. This means that if you search for gene expression in, for example, brain, besides searching for brain, the system also searches for substructures to brain in the hierarchy, such as rhombencephalon. If you enter a structure in the upper section of the form, then in the lower section, you must either enter another structure or check the box for "anywhere else." If you check the box for “anywhere else,” your search will return genes that have been shown to be expressed in the specified structure (or its substructures) but nowhere else. If you are unsure of the most appropriate structural term to use, browse the Mouse Developmental Anatomy Browser.
Developmental Stages	Use this field to select one or more Theiler stages (TS) to focus your search on a particular stage of embryonic development. The default for the lower box is Any stages not selected above. To select multiple stages or ages: On a Macintosh, hold the command (Apple, ⌘) key down and click on the desired stages; OR hold the shift key down and click on two stages to select those and all stages in between. On a PC, hold the control key down and click on the desired stages; OR hold the shift key down and click on two stages to select those and all stages in between. If you enter more than one stage in the upper box, for the query to return a particular gene, results demonstrating its expression must exist for at least one of the specified stages. If you enter more than one stage in the lower box, the query returns those genes for which there was an absence of expression in Any stages not selected above. Note: The Theiler system organizes development into stages defined by the appearance of specific developmental features. Embryos of the same gestational age can vary considerably with respect to development. Consequently, a Theiler stage does not precisely correspond to a particular age. For example, TS 21 applies to embryos between 12.5 and 14.0 days post conception (dpc), while TS 22 applies to embryos between 13.5 and 15.0 dpc. Therefore, if you are interested in all available records for embryos of age 14.0 dpc, you must select both TS 21 and TS 22. You can browse Stage descriptions for a list of the defining features for each Theiler stage.

The default tab for differential expression search results is the Tissue x Stage Matrix. This matrix is described in detail below. In brief,

• If the differential search queried for expression in a particular structure, that structure will be displayed on the upper row of the matrix; each cell in the row will be a shade of blue.
• If the differential search queried for lack of expression in a particular structure, that structure will be displayed on the lower row of the matrix. Red cells indicate that expression has been recorded as Absent in that structure; white cells indicate that expression has not been examined (or recorded) in that structure.
• If the "anywhere else box" was checked, only the structure where expression is detected is shown. To distinguish between the genes whose expression has been shown to be absent (compared to those whose expression has not been examined), click on the "View the Not Detected data for this gene set" filter below the matrix.
• The list of genes returned by the differential queries can be downloaded using the Export features found on the Genes tab.
• For any search, you can generate a list of genes whose expression has been shown to be absent (compared to those whose expression has not been examined) by using the "Detected?' filter, found above the summary, set to "No."

Top

What is the purpose of the Expression Profile Search?

It allows you to search for genes by their expression profile. You can define the expression profile of interest by specifying up to 10 anatomical structures and whether expression is present or absent in these structures. Examples of the questions you can ask are:

• Which genes are expressed in the metanephric mesenchyme but not in the metanephros epithelium and primitive collecting duct?
• Which genes are expressed in the photoreceptor layer but not in the retina ganglion cell and inner plexiform layers?

What values are acceptable in each of the Profile Search fields?

Anatomical Structure | Detected | Not Detected

Field	Description
Anatomical Structure	You can enter 1-10 anatomical structures. You can remove a structure by clicking on the X at the left of the row. The anatomical structures are taken from the Mouse Developmental Anatomy Browser. Begin typing a term of interest and an autocomplete list activates on the second character.
Detected	There must be at least one structure chosen as Detected (expressed) in any search. For the search to return a particular gene, results demonstrating its expression must exist for ALL structures chosen as Detected.
Not Detected	For the search to return a particular gene, either there are results showing it was Not Detected (absent) in the specified structure or there are no results, indicating its expression has not been analyzed or recorded in the database for that structure.

• RNA-Seq data are not yet included in these searches.
• The list of genes that fits the expression profile can be downloaded using the Export features found on the Genes tab.
• If a structure has conflicting expression results annotated to it, i.e., results showing both the presence and absence of expression of a particular gene, that gene is considered to be expressed. Expression results that are either Ambiguous or Not Specified are ignored by this query, i.e. they are treated as neither present nor absent.
• The search is limited to results from wild-type specimens. For the purposes of these queries, results from in situ reporter (knock-in) experiments will be considered wild type if the results are derived from a heterozygote.
• When applicable, you can use the Detected? = filter, found above the summary, to distinguish between genes whose expression has been shown to be absent (compared to those whose expression has not been examined). Selecting Detected? = No will limit the list on the Genes tab to those genes that have Not Detected results returned by the profile search. If more than one Not Detected structure was specified in the profile, you will need to use the Assay Results tab to distinguish which structure has been shown to lack expression for each gene.

Top

Can I perform batch searches for expression data with multiple genes?

Yes. There are 4 ways you can do this:

• In the Standard Search tab, you can enter a comma-separated list of genes.
• You can use the Batch Search tab to enter or upload a column of gene symbols or IDs. See Using the Gene Expression Batch Search for details.
• You can enter or upload symbols and IDs in the MGI Batch Query, though the results from this form lack the tissue-by-gene matrix view, multi-tabbed data summary and filtering options of the Gene Expression Data form results.
• You can search with a list of symbols or IDs in MouseMine and perform iterative refinement of results and link to other model organism Mines.

Top

Can I edit my original options and requery?

Yes, you can.

• If you want to edit a list or change output options:
  1. When your results appear, click the banner at the top of the page (Click to modify search). (The banner title changes to Click to hide search.) The window that appears contains your original search criteria.
  2. Make your changes.
  3. Click Search.

Notes:

• You can requery as many times as you like.
• If you had filtered your search results, when you modify your search, the filters are reset.
• Alternatively, click Reset to return to the original Gene Expression Data Query form.

Top

What are the advantages of the different results tabs?

The Genes, Assays and Assay Results tabs offer progressively detailed views of your expression search results. The 2 matrices provide interactive high-level overviews of temporal and spatial expression patterns and offer the ability to drill down to more detail.

Genes	The Genes tab provides a list of the genes that match your query along with the options to export your results as text files or to send the list to our Batch Query. Batch Query searches can associate the genes with additional information, such as functional or phenotypic data. See Using the MGI Batch Query.
Assays	The Assays tab presents a list of the unique assays that match your search, as well as their assayed genes and references.
Assay Results	The Assay Results tab provides the most detailed summary. It lists the structures assayed and whether or not expression was detected, in addition to the information provided in the 2 previous tabs.
Images	The Images tab allows you to quickly view images of expression results that match your search parameters.
Tissue x Stage Matrix	The Tissue x Stage Matrix tab offers a temporal and spatial overview of gene expression results. Toggles allow you to expand the anatomy hierarchy to reveal greater levels of detail in tissues of interest. This tab also allows you to filter results by tissue and developmental stage.
Tissue x Gene Matrix	The Tissue x Gene Matrix tab enables the comparison of expression patterns of different genes. As with the tissue x stage matrix, toggles allow you to expand the anatomy hierarchy to reveal greater levels of detail in tissues of interest. This tab also allows you to filter by gene.

Top

How do I interpret the query results summary page?

Your search results are displayed within six tabs, each noting the number of results: Genes, Assays, Assay Results, Images, Tissue x Stage Matrix and Tissue x Gene Matrix. With the Standard search, the Assay Results tab is selected by default. For Differential Expression searches, the Tissue x Gene Matrix tab is the default selection. See What are the advantages of the different results tabs? for a brief guide to choosing results tabs.

If your search returns more than 21 million assay results, only the first page of the assay results tab will be rendered. The other tabs will be blank. Such large search returns make our pages slow to load. To get full functionality in the summary, you will need modify your search or filter to return fewer results.

Genes

If you select the Genes tab, you are offered the ability to Export your results as a text, or forward the list of genes to the MGI Batch Query or MouseMine, where you can find additional information on the genes. Below the Export options, a table summarizes information on the gene(s). You can use the triangles in the column headers to sort by a column. The query summary contains the following fields:

MGI ID	MGI accession ID for the Gene
Gene	Symbol of the analyzed gene, linked to its MGI marker detail record
Gene Name	Name of the analyzed gene
Type	Marker feature type for the analyzed gene. See Genome Feature Type Definitions.
Chr	Chromosome to which the marker is assigned. Aside from a chromosome number, other possible values include MT (mitochondrial), XY (XY pseudoautosomal), UN (chromosome assignment is not known)
Genome Location-NCBI Build #	Genome coordinate range in base pairs and the build number of the C57BL/6J genome
cM	Cytogenetic position
Strand	Coding strand for the gene

Assay

If you select the Assays tab, the query summary contains the following fields:

Gene	Symbol of the analyzed gene for most assay types. Reflecting the nature of the assay, "Whole Genome" for RNA-Seq assays.
Assay Details	For most assay types, the word data linked to detailed expression results; the MGI accession ID of the assay (useful for searching purposes) appears in parentheses. For RNA-Seq assays, data set, followed by EBI's Expression Atlas set ID; clicking on the filter () icon will filter the summary for that data set.
Assay Type	Name of the assay type used (e.g., Immunohistochemistry, Northern blot, RNA in situ, and so on).
Reference	For most assay types, J number (MGI reference identifier) and short format citation for this reference linked to its MGI reference detail report. For RNA-Seq assays, the EBI Expression Atlas set ID and title, linked to the GXD RNA-Seq and Microarray Experiment Summary. There you will find details of the experiment and linkouts to the Expression Atlas (for access curated data and additional display and analysis tools), to ArrayExpress and GEO (to access the source data) and PubMed (to access the publication).

Assay Results

If you select Assay Results, you are offered the ability to Export your results as a text or export the RNA-Seq results to a heat map. Below the Export options, a table summarizes information on the assays. Ten columns are shown by default; an additional 4 columns are displayed when you use the “Show Additional Sample Data” toggle. You can use the triangles in the column headers to sort by a column. The query summary contains the following fields:

Gene	Symbol of the analyzed gene
Result Details	The word data links to detailed expression results in GXD (for most assay results) or linked to experiment detail pages at EBI’s Expression Atlas (for RNA-Seq assay results); the MGI accession ID of the assay or RNA-Seq data set id appears next in parentheses
Assay Type	Name of the assay type used (e.g., immunohistochemistry, Northern blot, RNA in situ)
Age	Age of the specimen used in the assay. The abbreviations for age annotations are as follows: E Embryonic day P Postnatal or postnatal day P w Postnatal week P m Postnatal month P y Postnatal year P adult Postnatal adult P newborn Postnatal newborn
Structure	Structure examined in the assay, given as a Theiler stage (TS) and anatomical structure name.
Cell Type	Cell type, if specified. Use of the Cell Type (CL) Ontology for annotation began in 2022, so there are not many annotations using it in the database.
Detected?	Was expression detected? Possible values are: Yes No Ambiguous Not Specified Note: No is used for RNA-Seq results when the TPM are less than 0.5. Not Specified is used when the authors do not report whether a gel band is present or absent. Ambiguous is used when the authors specify it or our curators cannot discern from the author's description whether expression is present or absent.
TPM Level (RNA-Seq)	Possible values are: Below Cutoff = 0.5 TPM Low=0.5-10 TPM Medium=11-1,000 TPM High=>1,000 TPM These mapping conventions are also used by EBI's Expression Atlas.
Biological Replicates (RNA-Seq)	Column is visible when the Show Additional Sample Data toggle is on. The number of biological replicates whose TPM were averaged to generate the TPM
Images	Name of the specimen or pane with associated images, linked to the appropriate section of the assay detail. Note: Images are shown only when we have permission from the publishers. There are no images for RNA-Seq results.
Mutant Allele(s)	If the specimen is a mutant, the allele pairs describing the mutant genotype appear in this field.
Strain	Column is visible when the Show Additional Sample Data toggle is on. Strain information for the specimen/sample, if specified. Use of involves: in the annotation indicates that the strains listed (separated by asterisks) contributed to the genetic background of the specimen; additional unspecified strains may also be part of the specimen's genetic background. Either: appears when the authors use more than one genetic background in a reference but do not specify which one is the background of the specimen.
Sex	Column is visible when the Show Additional Sample Data toggle is on. Possible values: Female, Male, and Pooled. When the field is blank, it indicates the author did not specific the sex.
Notes (RNA-Seq)	Column is visible when the Show Additional Sample Data toggle is on. Provides additional information about RNA-Seq samples, if needed.
Reference	For most assay types, J number (MGI reference identifier) and short format citation for this reference linked to its MGI reference detail report. For RNA-Seq assays, the EBI Expression Atlas set ID and title, linked to the GXD RNA-Seq and Microarray Experiment Summary. There you will find details of the experiment and linkouts to the Expression Atlas (for access curated data and additional display and analysis tools), to ArrayExpress and GEO (to access the source data) and PubMed (to access the publication).

Images

If you select the Images tab, the query summary contains the following fields:

Figure	A thumbnail image, linked to the MGI Image Detail page.
Gene	Symbol of the analyzed gene(s).
Assay Type	Name of the assay type(s) used (e.g., Immunohistochemistry, Northern blot, RNA in situ, and so on).
Specimen Type	Describes whether the sample is a whole mount, section, section from a whole mount, or optical section.
Result Details	Links to image and expression assay result details.

Tissue x Stage Matrix

The Tissue x Stage Matrix tab presents the query results in a grid view. The columns show developmental, Theiler, stages (TS) where expression was examined and the rows display the tissues that were examined. Theiler stages are defined using morphological criteria, not gestational age. For example, TS4 represents the blastocyst stage. For simplicity and ease of reference, the column headers also display the gestational age range over which each developmental stage may be found. The Theiler Stages are further defined here. The tissues consist of anatomy terms from The Mouse Developmental Anatomy Ontology. You can expand or collapse the anatomy hierarchy by clicking on the blue ► or ▼.

All expression results filters are applied across the six summary tabs, including those filters that originate in the matrix. On the matrix, check rows and/or columns and click the Filter () button to reduce the grid to just the rows and columns you have checked. Selected filters will be displayed above the tab; they can be removed by clicking on them. Any filtering done on the grid will apply to all the other data tabs as well. Individual cells have the ability to filter as well. Clicking on a cell displays the options, View All Results and View Images. Selecting one of those options filters the results and takes you to the appropriate summary tab where you can explore those results by visiting detail pages.

See What do the colors in a Matrix mean?

Tissue x Gene Matrix

The Tissue x Gene Matrix tab presents the query results in a grid view. The columns show genes whose expression was examined and the rows display the tissues that were examined. The tissues consist of anatomy terms from The Mouse Developmental Anatomy Ontology. You can expand or collapse the anatomy hierarchy by clicking on the blue ► or ▼.

In addition to the filters above the tabs, you can also filter by checking rows and/or columns in the grid and clicking on the Filter () button to reduce the grid to just the rows and columns you have checked. Selected filters will be displayed above the tab; they can be removed by clicking on them. Any filtering done on the grid will apply to all the other data tabs as well. Individual cells have the ability to filter as well. Clicking on a cell displays the options, View All Results and View Images. Selecting one of those options filters the results and takes you to the appropriate summary tab where you can explore those results by visiting detail pages.

See What do the colors in a Matrix mean?

Top

What do the colors in a Matrix mean?

The colors in the grid, blues for expression detected and reds for expression not detected, get progressively darker when there are more supporting annotations. The darker colors do not denote higher or lower levels of expression, just more evidence.

Grey bars indicate the only annotations for the tissue are ones for which the researchers examined the tissue but did not definitively state whether expression was present or absent, i.e. their description was ambiguous.

Squares with a red corner indicate that expression has been reported as being both present and absent in that tissue. Although this may be because different labs made conflicting observations, it may also be due to other factors, such as differences in the genotype or sex of the specimen, sensitivity of the assay used, or a difference in the transcript variant assayed.

A square with a gold corner indicates that although there are no annotations where expression was reported as being present in either the tissue or its substructures, there are expression results for substructures of the tissue that were reported as being either absent or ambiguous. See How are expression results propagated through the matrix anatomy hierarchy?

A circle in a square (found only in the Tissue x Stage Matrix) indicates the tissue exists at that Theiler stage but we have no expression data for it, whereas a blank square indicates that the tissue does not exist at that stage.

Clicking in any box in the grid will open a dialogue box. For blue squares, this box will indicate the number of annotations where expression was reported as being present in either the tissue or its substructures. It may also indicate additional expression results annotated to this tissue that were reported as being either absent or ambiguous; absent or ambiguous annotations to substructures will not be reported. See How are expression results propagated through the matrix anatomy hierarchy?

For red squares, this box will indicate the number of annotations where expression was reported as being absent in the tissue; annotations to substructures will not be reported. See How are expression results propagated through the matrix anatomy hierarchy?

For grey bars, this box will indicate the number of annotations to the tissue where the expression levels were reported in an ambiguous fashion. It may also indicate additional expression results annotated to this tissue that were reported as being absent. In neither case will absent and/or ambiguous annotations to substructures be reported. See How are expression results propagated through the matrix anatomy hierarchy?

Since squares with a gold corner indicate that all the expression results are absent or ambiguous annotations to substructures (rather than the structure represented by the square) the number of annotations is not indicated in the dialogue box.

In all cases, clicking "View All Results" will take you to the assay results tab where these results can be reviewed in detail. If there are annotations where expression was reported as being absent or ambiguous in a substructure, they will be shown on this tab (although they will not be included in the counts on the dialogue box). Clicking "View Images" will take you to the Images tab.

Top

How are expression results propagated through the matrix anatomy hierarchy?

The nature of the expression data captured by GXD requires present and absent expression results to be managed differently in the matrix anatomy hierarchy. From the statement, "expression was detected in the heart ventricle," we can infer that "expression is detected (somewhere) in the heart." In contrast, the statement "expression was not detected in the heart ventricle" does not imply that "expression is found (nowhere else) in the heart." Therefore, a positive expression result is displayed in the matrix as a blue square for that anatomical structure, but any parent structure will also display a blue square based on that present result. In contrast, absent expression results can only be represented for that structure in the matrix. Parent structures will not display a red square based on an absent result in a child structure.

Top

How do I filter my results?

You can filter results by attributes of the assay results or by vocabulary annotations to the genes analyzed. In the banner above your results, below Filter expression data by:, click the button for the desired filter. This will open a box displaying all the filter terms associated with the assay results or genes in your dataset. Check the boxes of interest and click Filter to to reduce your results to only those annotated to the checked items. Filters that have been applied are displayed below the buttons. You can remove them by individually clicking on them or by using the “Remove All Filters” button.

Note:

• You can apply as many filters as you desired.
• The filters are applied across all data tabs and to any file you export.
• Within one filter chosen terms are treated as "OR"s; between filters chosen terms are treated as "AND"s.
• Filtering out mutants will retain heterozygous in situ knockin mutants.

Assay Result Filters (left column)

Anatomical System	High-level terms to describe tissue analyzed
Theiler Stage	Developmental stage analyzed
Assay Type	Assay type used for study
Detected?	Either: Yes; No; Not Specified; or Ambiguous
TPM Level	For RNA-Seq results; either: Below Cutoff; Low; Medium; or High
Wild type?	Whether the specimen/sample was wild type or mutant.

Gene Filters (right column)

Gene Type	Type of marker, e.g. protein-coding, non-coding RNA gene
Molecular Function	High-level GO terms describing the molecular processes the gene product is involved with
Biological Process	High-level GO terms describing the biological processes the gene product is involved with
Cellular Component	High-level GO terms describing the cellular location of the gene product
Phenotype	High-level Mammalian Phenotype (MP) terms describing the phenotype of mice carrying mutant alleles of the gene
Disease	High-level Disease Ontology (DO) terms describing the diseases mice carrying mutant alleles of the gene manifest

Both Matrix tabs offer easy filtering by tissue. The Tissue x Stage Matrix also offers filtering by developmental stage/age, while the Tissue x Gene Matrix permits filtering by gene symbol. Check the rows or columns you wish to retain and then click the Filter () button in the grid to apply the filters. These filters are also applied to all tabs.

Top

Can I get my results as a tab-delimited text file?

Yes. The Genes and Assay Results tabs provide export options. Click on the Text File icon. Your web browser will download a file of the results. The Assay Results file will include all columns, regardless of whether or not the "Show Additional Sample Data" toggle is chosen.

Top

Can I create a heat map for my RNA-Seq results?

Yes. Click on the RNA Seq ► Heat Map button above the table of results in the Assay results tab. A browser window will open, displaying your selected data in Morpheus, a heat map visualization and analysis tool created at the Broad Institute. A summary of the data exported to Morpheus is at the top of the page. Each column of the heat map is a bioreplicate set; the GXD metadata annotations for the set are included in the column display. Each row in the heat map is a gene; the display includes its symbol and its MGI and Ensembl IDs.

Note: for any experiment, the number of samples displayed in GXD (and the heat map) may be fewer than the number in the Expression Atlas, the data source. The reasons for this are two fold:

1. In GXD we present the biological replicate sets from each experiment; these are derived by grouping samples that have the same source information (anatomical structure, developmental stage, mutated gene, strain and sex).
2. Experiments may contain some samples that are inconsistent with GXD's scope; data from these samples are not included in GXD.

Top

Can I send the results of my search to the MGI Batch Query or MouseMine?

Yes. Under the Genes tab you have the option of exporting your results as a text file, or sending the list of genes to the MGI Batch Query or MouseMine. If you click the Batch Query icon, the default Batch Summary will appear for your list of genes. Use the Click to modify search button to expand your search results to include additional information such as Gene Ontology (GO) terms, Mammalian Phenotype (MP) terms or Human Diseases associated with the genes. MouseMine offers flexible querying, numerous predefined query templates, iterative refinement of results, and linking to other model organism Mines.

See also Using the MGI Batch Query.

Top

Why is there a link to Gene Expression Literature records in my query summary?

At the end of the query summary, there may be a link to Gene Expression Literature content records matching the parameters of your detailed expression data search.

• The first step of annotation in GXD is to create literature content records containing information on the genes and ages analyzed and assay types used in each reference. All expression papers have content records.
• The next step is for the curators to extract the detailed expression data from these references. The types of data annotated at this stage include: the anatomical structures analyzed; the authors' description of the pattern and strength of expression; and the genetic origin of the sample. Because annotation process is time-consuming, not all expression references have detailed data in the database.

Thus, to ensure that your query returns all the references of interest to you, searches of the detailed expression data also include searches of the literature content records (whenever query parameters apply to both these types of data). Filtering your query summary will not affect the number of literature records returned, whereas modifying your query will.

Which expression data query parameters generate an expression literature search? Which do not?

The following detailed expression data query parameters generate an expression literature search:

• gene symbol/name
• function, phenotype or disease term
• developmental stage
• assay type.

Because of the limited annotation of content records, the following parameters do not generate an expression literature search:

• anatomical structure
• detected in (expression present)
• not detected in (expression absent)
• mutated gene symbol/name.

Note: You can query gene expression literature content records directly using the Gene Expression Literature Query Form.

Top

How often are annotations added?

GXD curators add new annotations from the literature every day. The MGI web site is updated with these and other new data once a week.

Top

Are there examples of query results?

The following examples list the field values to set on the Gene Expression Data Standard Query to perform each query. Leave the default values in all other fields.

1. When and where is En1 expressed?

   Genes…Search for expression data for…: en1
   

   This query returns a list of results from assays examining En1 expression. This summary report includes the gene symbol, result details linked to an assay record (click data), the assay type used, the anatomical system assayed, the age of the specimen, the specific structure examined, an indication of whether or not expression was detected, links to images, the mutant allele pairs describing the mutant genotype of the specimen (if applicable), and the reference from which the data were curated, linked to its abstract.

2. What genes are expressed in the heart at E8.5 in wild-type mice only?

   Expression: detected in
   Anatomical Structures: heart
   Developmental Stages (dpc): Select the Ages (dpc) tab and then E8.5
   Mutant / wild type:   Wild type specimens only
   

   This query returns a list of assay results for genes expressed in the heart at E8.5. Click the Genes tab to view a summary report that includes the gene symbols (with links to MGI detail pages), gene types and genomic map coordinates. The report does not include any expression results obtained from mutant alleles. Note: Each developmental stage in the example includes coverage of embryonic day 8.5. See Theiler System for details.

3. What members of the Wnt gene family are expressed in TS 16?

   Genes…Search for expression data for…:wnt*
   Expression: detected in
   Developmental Stages (dpc): TS 16
   

   This query returns a list of assay results for Wnt family genes whose expression was detected in Theiler Stage 16.

4. When and where are genes involved in axon guidance expressed?

   Genes…A set of genes defined by: axon guidance
   

   This query returns a list of all assay results for all genes known to be involved in mouse axon guidance.

5. What genes have been shown NOT to be expressed in either the diencephalon or any structure within the diencephalon?

   Expression: not detected in
   Anatomical Structure(s): diencephalon
   

   This query returns a list of genes found not to be expressed in the diencephalon or in anatomical structures within the diencephalon, such as the hypothalamus, 3rd ventricle, or pituitary gland.

6. What genes involved in cell adhesion are expressed in stomach mesenchyme?

   Genes...A set of genes defined by: cell adhesion
   Expression: detected in
   Anatomical Structures: stomach mesenchyme
   

   This query returns a list of genes involved in cell adhesion found to be expressed in the stomach mesenchyme.

Top