diff --git a/Citations.md b/Citations.md index a4e4dea..8fc18f5 100644 --- a/Citations.md +++ b/Citations.md @@ -17,10 +17,10 @@ For the collected cells, biomarker (marker) labels were assigned as either posit 3. as a custom name provided by the researchers. For a full table of the naming conventions, please see the Figure X. ### Spatial-Interaction Tool -[In Prep] +*In Prep* ### Neighborhood Profiles -With the dataset phenotypes set, +Cell densities were calculated by finding the per-phenotype counts of cells surrounding each cell in the dataset divided by the area of the annuli surrounding concentric radii around each of those cells. Spatial UMAP methods were performed using the *umap* package. ## Citation Weisman, Andrew; Smith, Dante. (2023). multiplex-analysis-web-apps. GitHub. https://github.com/ncats/multiplex-analysis-web-apps diff --git a/assets/images/MaG_Column_Filter.png b/assets/images/MaG_Column_Filter.png new file mode 100644 index 0000000..51deb88 Binary files /dev/null and b/assets/images/MaG_Column_Filter.png differ diff --git a/assets/images/MaG_CurrentPheno.png b/assets/images/MaG_CurrentPheno.png new file mode 100644 index 0000000..2f02f16 Binary files /dev/null and b/assets/images/MaG_CurrentPheno.png differ diff --git a/assets/images/MaG_PhenoAssign1.png b/assets/images/MaG_PhenoAssign1.png new file mode 100644 index 0000000..71b843c Binary files /dev/null and b/assets/images/MaG_PhenoAssign1.png differ diff --git a/assets/images/MaG_PhenoAssign2.png b/assets/images/MaG_PhenoAssign2.png new file mode 100644 index 0000000..175ce46 Binary files /dev/null and b/assets/images/MaG_PhenoAssign2.png differ diff --git a/mutliaxial_gating.md b/mutliaxial_gating.md index d1f44ce..219e625 100644 --- a/mutliaxial_gating.md +++ b/mutliaxial_gating.md @@ -11,8 +11,34 @@ After importing your dataset in the `Data Import and Export`, you should be able ![](./assets/images/MaG_LoadData.png) -## Investigating Features +## Full App Loads Once the data is loaded, the full `Multi-axial Gating` app is revealed to be the following four sections -1. +1. Column Filtering +1. Current Phenotype +1. Phenotype Assignments +1. New Dataset -![](./assets/images/MaG_FullApp.png) \ No newline at end of file +![](./assets/images/MaG_FullApp.png) + +## 1. Column Filtering +The Column Filtering panel is the first place to begin identifying the feature or features that will make up your sub-phenotypes. All columns present in the dataset are populated in this select box. Selecting any one feature plots the range of values in the histogram below. For continuous variables such as florescence intensity, the historgram is smoothed using a kernel density estimation (KDE). For categorical variables, there is no smoothing. Above the plot is a range slider, which when first loaded is set to the min and maximum values of the histogram. By moving each end of the range slide, you can set a new range of values to consider a given variable. The histogram will update the selected legend entry based on the values of the range slider. Once you have selected a target range for a variable, click the button below titled *Add Column Filter to current phenotype*. This will create an entry in the next section, Current Phenotype, with the selected marker and the range of values selected. Once a marker is added to the Current Phenotype section, it is no longer selectable in the Column Filter select box. If at any point you want to re-select a range for a given phenotype, this can be done in the next step. + +![](./assets/images/MaG_Column_Filter.png) + +## 2. Current Phenotype +A given phenotype can be made up of multiple marker ranges, but only one instance of each marker. As marker ranges are selected in the Column Filter step, they populate in the table describing the Current Phenotype. Each field of this dataframe is editable, so if you would like to more preciely tune the threshold values for min and max, value that is possible to do. If a whole row has become unnecessary for the current Phenotype definition, you can select the cell to the left of the row, and click the delete button in the top right of the dataframe. If at any point, you want to completely start over the current phenotype definition, you can do so by clicking the *Reset Data Editor* button. + +Once the list of markers and their ranges has been selected, you can name the phenotype in the text box labeled Phenotype name. After naming the phenotype, click the *Add phenotype to assignments table* button. After doing so the table in this section will clear, and the table in the next section 'Phenotype Assignments' will populate with the name of the newly customized phenotype, and the marker min and max values used to define the phenotype. In addition, after adding the new phenotype definition to the Phenotype Assignments table, the list of Features in the Column FIlter drop down box will repopulate with all available column names from the original dataset. + +Repeat the above instructions in Sections 1. and 2. as many times as are required for your study definitions. Once all phenotypes have been defined, move on to section 3. + +![](./assets/images/MaG_CurrentPheno.png) + +## 3. Phenotype Assignments +Once a phenotype is defined in Section 2, they will be added to the Phenotype Assignments table in Section 3. The table details the name of each phenotype in the first column, and the subsequent columns are the min and max values of the markers listed for a given phenotypes. Like the previous table, each cell is editable if finer grain control is desired for a given field. If there is a need to reset the whole table, this can be done by clicking the *Reset data editor* button. + +Once the Assigned Phenotype definitions table is complete, its time to integrate it back into the input dataset. This is done by clicking the *Generate new dataset from the phenotype assignments* button. + +![](./assets/images/MaG_PhenoAssign1.png) + +![](./assets/images/MaG_PhenoAssign2.png) \ No newline at end of file