...

Classification in Partek Flow

The Garnett[¹] automated cell type classification algorithm has been wrapped into Partek Flow as the Classification task in Partek Flow. Just like . As with the original Garnett tool, Classification in Partek Flow works on single-cell data, along with a cell type definition (marker) file, and trains a regression-based classifier. Once a classifier is obtained and published, it can be applied to classify future datasets from similar tissues. To improve the user experience, both maker the marker file (.txt) and the classifier file (.rds) have been implemented as library files in Partek Flow. 

Garnett classifiers in Partek Flow

Partek Flow hosts a few selection of pre-trained classifiers as Managed classifiers. The list of available classifiers can be found here[²]. If a managed classifier exists for your data type, we recommend that you try it. Besides In addition to the Managed classifiers, the Project classifiers trained on the same dataset prior to your classification can , may also be used to classify cell type. Project classifiers can be promoted to Managed classifiers if users publish them. 

Classify cell type task in Partek Flow

To classify cell type using a pre-trained classifier:  

Select any non-normalized single cell data node, Filter counts here which contains . Filtered counts, which contains the raw count, is used here. Next, choose the Classify cell type task in the Classification section (Figure 1)

...

If using the Managed classifiers users tool for the first time in Partek Flow, you will be asked to create a new classifier file if it is the first time to run the task in Flow (Figure 2a). Users could may select either the Download Garnett classifier that matches the species and tissue type with the their dataset working on from a Partek Flow maintaining maintained list or the Import Garnett classifier that’s that is trained out of Partek Flow (Figure 2b). Then Next, push the Create button to create the classifier file. Once the right correct classifier file has been created, Select select Finish to start running the task (Figure 2d). 

...

If users would like to classify cell type with the classifier obtained from the same project, ‘Project classifiers’ needs to be picked up selected from the dropdown list. Next, then use ‘Select data node’ to choose the classifier before clicking the Finish button (Figure 3). 

...

No matter Regardless of which type of classifier was applied, it Partek Flow will output a new data node named Classify result (Figure 4). The These outputs of cell type annotation are exactly the same as Garnett[¹]. Downstream analysis like tasks such as normalization , and PCA , etc can be run performed on the Classify result data node.

Train classifier in Partek Flow

To train a classifier with a list of biomarkers with your own dataset:

Select any non-normalized single cell data node, Filter Filtered counts is used here. Next, choose the Train classifier task in the Classification section (Figure 1)

SimilarlySimilar to previously described steps, first time users will be asked to create the Marker file . Partek Flow does not currently doesn’t host any marker files, however, users can may add them as library files.   Marker file files should be a .txt file with the marker information in correct format. The same example in the Garnett tutorial of a simple valid Marker file is provided here (Figure 5).  For more details on how to construct a Marker file, please refer to Garnett tutorial[³]. Click Next, click the Create button , and Partek Flow will then save the file with the name that users provided provide for future use (Figure 6a).

...

Train classifier task report in Partek Flow

Once the task has been finished, click the Classifier datanode data node and choose the Task report in the Task results section, or prompt it to be one of the Managed classifiers by clicking the Publish classifier task in the Classification section (Figure 7).

...

Users will find two parts in the task report: the marker evaluation plot (Figure 8) and the classification gene table (Figure 9).  The marker evaluation plot provides some key information about whether the chosen markers are goodoptimal. Ambiguity scores are calculated for each of the markers which indicates how many cells receive ambiguous labels when this marker is included. The classification gene table could may give a hint to which genes are chosen as the relevant genes for distinguishing between different cell types. 

...

Other parameters adjustable parameters in this task that you can adjust include:

Number of Unknown: it this tells Garnett how many outgroup cells it should compare against. The ; the default is 500. For a dataset with fewer cells, the number should be smaller.

...