S.No.	Software Name	Web Address	Publicly Available	Architecture Details/Flowchart	Publication for Developed Database	Min System Requirement	Other Details
1	DEELIG	https://github.com/asadahmedtech/DEELIG	Yes	http://14.139.62.220/software_uploads/325.jpeg	http://14.139.62.220/published_databases/326.pdf	Linux System with NVIDIA GPU	We propose a deep-learning-based approach to predict ligand (eg, drug)—target-binding affinity using only structures of target protein (PDB format) and ligand (SDF format) as inputs. Convolutional neural networks were used to learn representations from the features extracted from these inputs and the hidden layers in the affinity prediction task. We used 2 approaches for feature extraction—atomic level as well as the composite level and compared their performance using the same network. We have trained on complexes from PDB across all taxa filtered as per a few starting criteria including crystal quality.
2	DSDBASE2.0	http://caps.ncbs.res.in/dsdbase2/	Yes	http://14.139.62.220/software_uploads/328.png	http://14.139.62.220/published_databases/329.pdf	No system requirement	To the best of our knowledge, DSDBASE is the first database of its kind that organizes all disulphide bonds of proteins in one platform. A total of 153,944 PDB entries, 216,096 native and 20,153,850 modelled disulphide bond segments from PDB January 2021 release have been included in the updated DSDBASE2.0 database. Furthermore, the current database also provides tools for a user-friendly search of multiple disulphide bond containing loops, along with their functional annotation using GO and subcellular localization of the query. Additionally, an independent algorithm - RANMOD has been added as a useful tool into the webserver. By the very nature of the RANMOD procedure, with constraints on disulphide bond connectivity, multiple conformations are proposed as 3D models of the polypeptide.
3	PASS2.7	http://caps.ncbs.res.in/pass2/	Yes	http://14.139.62.220/software_uploads/330.jpeg	http://14.139.62.220/published_databases/331.pdf	No system requirement	In the latest version of the PASS2 database, PASS2.7, a total of 2024 superfamilies have been considered, of which 1219 are MMSs and 805 are SMSs. These superfamilies contained a total of 14?323 SCOPe domains. SCOPe groups superfamilies into 12 classes, of which 7 are available for download from the ASTRAL compendium. The current update of PASS2 (i.e. PASS2.7) is following the latest release of SCOPe (2.07) and we provide data for 14?323 protein domains that are <40% identical and are organized into 2024 superfamilies. Several useful features derived from the alignments, such as conserved secondary structural motifs, HMMs and residues conserved across the superfamily, are also reported. Protein domains that are deviant from the rest of the members of a superfamily may compromise the quality of the alignment, and we found this to be the case in ?7% of the total superfamilies we considered. To improve the alignment by objectively identifying such ‘outliers’, in this update, we have us