This site  supplements the original DTC Lab Software Tools site http://teqip.jdvu.ac.in/QSAR_Tools/DTCLab/ 

This site will host new software tools developed in 2021 and onwards

IMPORTANT NOTICE (15 Feb 2023): ALL EXISTING AND NEW DTC LAB TOOL USERS ARE REQUESTED TO SIGN THE NEW LICENSE AGREEMENT FORM (DATED 15 FEBRUARY 2023) TO RE-VALIDATE THEIR LICENSE AND SEND IT TO THE EMAIL ADDRESS GIVEN AT THE BOTTOM OF THIS PAGE.  The Licensee will fill in the form https://forms.gle/1r3TTy7RmZCQvqBt5 (or any other as recommended at the time work in the DTC Lab websites) for every research problem/paper where the tool/tools is/are used.

Quantitative Read-across

Read-across is a technique for predicting endpoint information for one substance (target substance), by using data from the same endpoint from (an)other substance(s), (source substance(s)) ((https://echa.europa.eu)

This software predicts toxicity (biological activity, in general) of nanomaterials (chemicals, in general) using different similarity based functions and also evaluates the quality of predictions in terms of different validation metrics like q2ext_F1  q2ext_F2 and rmsep, if experimental data of query compounds are available.

Environ Sci: Nano 9, 2022, 189-203  (Tutorial Video Playlist)

OpenTox Abstract

RASAR

RASAR combines read-across (unsupervised) with QSAR (supervised) techniques. 

This tool quickly computes various similarity-based descriptors which can be used to generate q-RASAR models. Prior to using this tool, the optimized setting of the hyper-parameters should be obtained using Read-Across-v4.1 tool, and this setting should be employed to calculate the similarity-based descriptors using RASAR-Desc-Calc-v3.0.1. The user may additionally choose to employ structural and physicochemical descriptors along with the similarity-based descriptors for q-RASAR model development. 

References

[1] Banerjee A., Roy K., 2022. First report of q-RASAR modeling towards an approach of easy interpretability and efficient transferability. Mol. Divers. 26, 2847-2862, DOI: 10.1007/s11030-022-10478-6

[2] Banerjee A., Chatterjee M., De P., Roy K., 2022. Quantitative predictions from chemical read-across and their confidence measure. Chemom. Intell. Lab. Syst. 227, 104613. DOI: 10.1016/j.chemolab.2022.104613

Auto RA Optimzer

This tool is a fully automated optimizer of the hyperparameters (σ, γ and the no. of close source/training compounds) associated with the Read-Across-based predictions. The selection of hyperparameters should be based on the performance of the model on a validation (or sub-test) set and not on the test set.  

Reference 

Chatterjee M., Banerjee A., De P., Gajewicz A., Roy K., 2022. A novel quantitative read-across tool designed purposefully to fill the existing gaps in nanosafety data. Environ. Sci: Nano 9, 189-203. DOI: 10.1039/D1EN00725D [Content free to download from the Publisher site]

Download Link v1.0 (Restricted) (Release date 18 September 2022)

To use the software,  please fill in the license agreement form (mandatory) and email it to the following email ID. Java must be installed on your computer to run the program.

Other Small Tool(s)

Mixture Descriptor Calculator 

Mixture Descriptor Calculator quickly computes mixture descriptors using a weighted descriptor generation approach (Chatterjee M, Roy K,  Journal of Hazardous Materials 408 (2021) 124936, https://doi.org/10.1016/j.jhazmat.2020.124936 ).

This tool is compatible with Windows x86 (32 bit).

Download link v 1.0 (Restricted) Release date December 03, 2021

Download request link 

To get access to the software,  please fill in the above form (mandatory) and click on the download link. The link should normally be accessible within 24 hours if granted. 

Mixture_Desc_Calc v1.0

This tool quickly computes the mixture descriptors based on the fraction of the individual components specified by the user. It utilizes three different mathematical expressions for calculation of the mixture descriptors, and the results are printed on separate sheets of the output file.

Reference

Chatterjee M, Roy K, J Hazard Mater 408, 2021, 124936, https://doi.org/10.1016/j.jhazmat.2020.124936

Download link v1.0 (Release date 28.11.2022) (Restricted)

To use the software,  please fill in the license agreement form (mandatory) and email it to the following  email ID. Java must be installed on your computer to run the program.

Klassification v1.0

This simple tool quickly computes Classification-based validation metrics from given observed and predicted values  (Reference: Roy K, Kar S, Das RN, Understanding the Basics of QSAR for Applications in Pharmaceutical Sciences and Risk Assessment, Academic Press, NY, 2015, https://doi.org/10.1016/C2014-0-00286-9)     

Reference

Banerjee A, De P, Kumar V, Kar S, Roy K, Chemosphere, 2022, no. 136579, https://doi.org/10.1016/j.chemosphere.2022.136579 

Download link v1.0 (Release date 16.02.2022) (Restricted)

To use the software,  please fill in the license agreement form (mandatory) and email it to the following  email ID. Java must be installed on your computer to run the program.

Molecular Similarity Calculator v1.0

This tool quickly computes the similarities between compounds constituting the dataset. The similarity values are computed based on the Euclidean Distance (relative similarity scaled between 0 and 1), Gaussian Kernel similarity and Laplacian Kernel similarity (absolute similarity). Using the computed similarity values, a heat map can be constructed in Excel to detect the possible outliers (compounds having very low level of similarity with most other compounds).

Reference

Banerjee A, De P, Kumar V, Kar S, Roy K, Chemosphere, 2022, no. 136579, https://doi.org/10.1016/j.chemosphere.2022.136579 

Download link v1.0 (Release date 06.07.2022) (Restricted)

To use the software,  please fill in the license agreement form (mandatory) and email it to the following  email ID. Java must be installed on your computer to run the program.

Leverage (Hi) Calculator

This tool quickly computes the leverages of the training and test set compounds when the descriptor matrices are provided as input. The leverage value is an important diagnostic for the detection an structural outlier in the training test and a compound outside the applicability domain in the test set. 

Download link v2.0 (Release date 05.01.2023) (Unrestricted)

Features in this upgrade: This tool now calculates the leverages of the training and test set compounds along with their AD status (Outlier in case of the training compounds and Outside AD in case of the test compounds) with respect to the h* value.

Download link v1.0 (Release date 18.12.2022) (Unrestricted)

To use the software,  please fill in the license agreement form (mandatory) and email it to the following email ID. Java must be installed on your computer to run the program.

Scale v1.0

This tool quickly calculates the standardized descriptor values for the training and test set compounds using corresponding training set descriptor means and standard deviations. This might be used before QSAR modeling to work with standardized descriptor values.

Reference: George W. Snecdecor, William G. Cochran, Statistical Methods, Wiley, 1989

Download link v1.0 (Uploaded on 26 March 2023) 

To use the software,  please fill in the license agreement form (mandatory) and email it to the following email ID. Java must be installed on your computer to run the program.

MDF-Identifier-v1.0

This tool quickly computes the most discriminating set of features from an input training set. This is a feature selection tool where the most discriminating features are identified which can be used for (mainly, classification-based) model development. However, it may also be useful for the dimensionality reduction of the feature matrix in case of quantitative response data.

Reference:

Nandy, A.; Kar, S.; Roy, K. Molecular Simulation 2014, 40, 261-274

Download link v1.0 (Uploaded on 11 May 2023) 

To use the software,  please fill in the license agreement form (mandatory) and email it to the following email ID. Java must be installed on your computer to run the program.

DTC Virtual Sample Generator v1.0

This tool quickly computes virtual data points (samples) generated from a given dataset using a Gaussian Kernel similarity-based approach. 

Reference

1) Sutojo, T.; Rustad, S.; Akrom, M.; Syukur, A.; Shidik, G, F.; Dipojono, H. K. A machine learning approach for corrosion small datasets. npj Mater. Degrad. 7, 2023. 

Download link v1.0 (Restricted; Uploaded on 13 August 2023) 

To use the software,  please fill in the license agreement form (mandatory) and email it to the following email ID. Java must be installed on your computer to run the program.

Prediction Tool(s)

hERG_Toxicity_Calculator v2.0

UPDATE in this release: This tool now additionally detects structural outliers by computing the leverages of the external set compounds and checks whether a compound lies inside or outside the applicability domain based on the h* value (Gramatica, 2013, In: Computational Toxicology. Methods in Molecular Biology, B, Reisfeld, and A. Mayeno, eds., vol 930. Humana Press, Totowa, NJ. doi: 10.1007/978-1-62703-059-5-21)

This tool quickly provides the quantitative prediction of the potential cardiotoxicity induced by a compound by interacting with the hERG K+ channel using a q-RASAR model developed by the DTC Laboratory [1].  The q-RASAR approach has been described previously in [2] and [3].

[1]    Banerjee, A.; Roy, K. 2023, Chemom Intell Lab Syst, https://doi.org/10.1016/j.chemolab.2023.104829. 

[2]    Banerjee, A.; Roy, K. 2022, Mol. Divers. 26, 2847-2862

[3]    Banerjee, A.; De, P.; Kumar, V.; Kar, S.; Roy, K. 2022. Chemosphere 309, 136579

Disclaimer: Predictions to be used for research purposes only

Download v2.0 (Unrestricted from 20.04.2023) (Release date February 06, 2023)

Download  v1.0 (Restricted) (Release date February 02, 2023)

To use the software,  please fill in the license agreement form (mandatory) and email it to the following  email ID. Java must be installed on your computer to run the program.

Skin Sensitizer Calculator v1.0

This tool quickly computes the quantitative skin sensitization potential of query chemical(s) in terms of pEC3 using a PLS q-RASAR model and states whether a particular query compound is toxic, non-toxic, or borderline. It also checks the AD status of the query compound(s) using the leverage approach and identifies the outliers.

Disclaimer: Predictions to be used for research purposes only

Reference: Banerjee A, Roy K, Environ Sci: Process Impacts, 2023, https://doi.org/10.1039/D3EM00322A 

Download  v1.0 (Unrestricted from 10 Sept 2023) (Release date July 07, 2023)

To use the software,  please fill in the license agreement form (mandatory) and email it to the following  email ID. Java must be installed on your computer to run the program.

Visit the Original site

Contact Prof. Kunal Roy kunal.roy@jadavpuruniversity.in 

Last update: October 29, 2023