This tool calculates the novel ARKA descriptors – an innovation from the DTC Laboratory. This framework presents a novel supervised dimensionality reduction technique that may be useful for the data set modelability analysis, the identification of activity cliffs, and small dataset classification modeling.

Version 2.1: Download from here (uploaded on October 08, 2024; unrestricted from December 1, 2024)

[Features available in this upgrade: 1) For ease of use, this tool now specifies whether a compound is an activity cliff or not. 2) The positive and negative compounds are mentioned based on the training set mean response (in the case of the quantitative response data).]

------------------------------------------------------------------------------------------------------------------------------------------

Previous versions:

Version 2: Download from here (uploaded on August 19, 2024; presently restricted)

[Features available in this upgrade: 1) Identification of the quadrant where a particular data point lies in the ARKA_2 v/s ARKA_1 plot. 2) Computation of the Euclidean Distances of the data points from the origin of the ARKA_2 v/s ARKA_1 plot. The first analysis helps to identify the activity cliffs and data points having good modelability from their position in the plot (quadrant), while the second analysis signifies the confidence – the greater the Euclidean Distance, the greater the confidence.] 

Version 1: Download from here (uploaded to this site on March 25, 2024; Presently restricted)

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------

Ref: Banerjee, A., Roy, K., 2024. ARKA: A framework of dimensionality reduction for machine-learning classification modeling, risk assessment, and data-gap filling of sparse environmental toxicity data, Environ Sci: Process Impacts, 26, 991-1107, https://doi.org/10.1039/D4EM00173G .

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------

To use this tool, please fill in https://forms.gle/1r3TTy7RmZCQvqBt5 and sign the License agreement form

Further updates on ARKA from here