On the Nygen Analytics platform, you can add metadata such as dataset labels and cell annotations to your dataset.

Dataset labels are the metadata details for each file you upload or dataset received. This can include any information about the dataset, for example, which technology was used to create this data, the sample name or treatment condition this dataset was from. When you merge datasets, the dataset labels will become cell annotations in the new merged dataset. This keeps track of the origins as well as the details of the cells in the merged dataset.

Cell annotations can be any data on a cell or groups of cells. These annotations are at cell level, meaning each cell will have a label in a group/category. For example, in a dataset you can have cell type annotations, each cell is labelled with a cell type such as T-cell, monocytes, epithelial cell, etc.

In this section we will cover on how to:

1. Create new labels for your dataset
   1. Adding metadata labels to a single dataset
   2. Metadata table for multiple datasets
2. Import cell annotations to your dataset
   1. How to import cell annotations
   2. Where to find your imported data
3. Import an existing analysis

‍

Create new labels for your dataset

Adding metadata labels to a single dataset

‍


Steps

1. From your projects page, click on the name of your dataset to open the details for that dataset.
2. Use the ‘+ Add’ button from the Metadata labels section, this will let you create new field and add a value for this dataset. For example, the field can be treatment condition or sample number, and the corresponding value, ‘disease’ or ‘sample_01’.
3. A pop-up will show you when the label is added successfully and then you can add more field if needed.

‍

Metadata table for multiple datasets

You can add metadata to your datasets using a metatable or by uploading a csv file.

‍


Steps

1. When you have uploaded or received multiple files, you can utilise the metadata table option from the projects page to add labels.
2. Here you will find a table where you can choose to add new labels or import a csv file with the dataset labels for each file. Importing a csv is recommended if you have a large number of files.
3. A .csv file contains a table of values separated by commas in plain text. We expect a column for id (dataset name) and at least one label column (for example: sample, condition)
4. Save the changes made to the table, and now you can start a new analysis or merge your datasets.

‍

‍

Importing cell annotations to your dataset

How to import cell annotations

Here are some simple steps you can follow for importing cell annotations.

‍

Steps

1. From the explorer page for your dataset, you may find the Import button from the Cell metadata dropdown menu.
2. This will open the Import features panel where you can open your file browser for upload and some details on what we expect in the file. For cell annotations, we require:
   • a comma-separated table, CSV format file
   • a barcode column which should contain the same barcodes or cell index in your dataset, this column is required to be named as ‘barcode’. You can still import a file if it contains extra cell data with barcodes that are not in the dataset or if some cells are missing.
   • feature columns, these columns can be of any name (e.g. sample_name, cell_type, treatment) and have text or even numerical values.
3. You will also have the option to choose which columns from your csv to upload. You can uncheck the boxes for the columns you wish to exclude.
4. Once confirmed, hit Import! You will also see details of your data when you have successfully imported the file.
5. If you encounter issues importing, please get in touch with our support team!

‍

Where to find your imported data

‍


Steps

1. Once you have successfully added data, you may find the imported cell annotations from the Group by dropdown menu. This is where you find the groups for example, cell types, treatment conditions, sample name, etc and the categories within each group.
2. You can also remove any of group of cell annotations that you have imported.
3. To see the cell annotations, click on the Edit icon next to the Group by.
4. Cell metadata in numeric values, e.g. number of genes per cell from another analysis, are stored under Cell metadata.

‍

‍

Importing an existing analysis

Steps

1. From your projects page, click on the ‘…’ for more options for the dataset, select ‘Import analysis’.
2. You can import an existing analysis for your dataset onto the platform in the form of a csv file, which contains tabular data separated by commas. Each row in the csv should represent a cell in your dataset. We use the barcode column to match the table data to your dataset. Here are some details on the columns we expect in the csv file:
   • barcode - We require the same cell barcode/identifier as the ones in your dataset
   • cluster - Cell cluster labels to identify which cluster each cell belongs too
   • x and y - These are the x and y coordinates for the main scatter plot (e.g. UMAP coordinates)
   • my column - You can include more columns for cell annotations such as cell types, sample, etc
3. Upload your analysis file and once the upload is complete, you can find your details on your analysis from the ‘Available analysis’ panel.
4. You can edit the name of the newly imported analysis.
5. If the analysis csv file fails to import, please get in touch with our support team!

‍

💡Tips

• Add as much metadata as you can to your datasets before merging. Why? Metadata added to a dataset will carry over to any new merged dataset. This can be very useful in identifying the source datasets of cells or samples in your downstream analysis.
• Forgot to upload/add metadata from the Upload page? You can still add them to your datasets from your projects page, the steps taken for the metadata table are exactly the same as you would do from the Upload page.
• If you have a large number of datasets, it will be easier to add metadata to the files with a csv file.

‍