This is software to automatically calculate the total number of hydrogen bonds, salt bridges, and amino acids in large numbers of proteins. The software takes either UniProt .fasta or protein database .pdb files as input.

If starting with UniProt .fasta files:

• Upload a .fasta file (downloaded from the UniProt database (http://www.uniprot.org/)
Indexing of 10K+ files require separate process. After indexing process is done, please, reload the page to see results.
• Once files have been uploaded, click “Calculate using .fasta” button to calculate hydrogen bonds and salt bridges for each protein.
• this will first use SWISS MODEL algorithm (https://swissmodel.expasy.org/interactive) to generate a protein database (.pdb) file for each protein sequence in the .fasta file. SWISS MODEL will be unable to model proteins with an “X” in the amino acid sequence or if it's shorter then 30.
• species, protein name, gene name, length, and QMEAN values are extracted from the resulting .pdb file
• modeling is restricted to 2000 protein queries per 24 hours. A bar will show progress and the results table will be updated after each protein has finished.
note that .pdb files will not be created for every protein on the input list, only proteins for which SWISS MODEL has a suitable template in their database
• QMEAN scores below -4.0 are an indication of a low quality model
• hydrogen bond prediction uses HBOND algorithm
• salt bridge prediction uses ESBRI algorithm
• Data can be downloaded as .TXT, .CSV, or .XLSX using “Download” button

If starting with .pdb files:

• Drag or upload Swiss Index file for .pdb cross reference to UniProts. Drag or upload multiple .pdb files. Click "Calculate using .pdb" to begin calculating hydrogen bonds and salt bridges for each file.
• in cases where multiple .pdb files have been generated for the same protein (determined by duplicate UniProt IDs), only the model with the highest QMEAN score will be added to the results table.
• for this reason, the number of input .pdb files may not match the number of results (i.e rows) in the data table. Also same sequence will have same best .pdb file for it, but it will have different UniProt ids for different species.
• if SWISS MODEL bulk modeling approach was used, an index file can be uploaded which will cross-reference protein and gene names if present in the index.
• Data can be downloaded as .TXT, .CSV, or .XLSX using “Download” buttons.

If counting amino acids:

• Drag or upload .fasta file (downloaded from the UniProt database (http://www.uniprot.org/)
• Click the “Count Amino Acids” Buttons to start the calculations
• Amino acid counts are automatically transferred to a separate results table. Use the “Amino Acids Results Table” link to access the data.

Accessing the database:

• Structural data compiled through this software is stored in a database. Use the links below to access the database.
• INDEX displays data from the index files provided with SWISS MODEL bulk protein modeling approach
• UNIPROTS displays data sorted by unique UniProt identifier
• SPECIES lists species and number of proteins from each species in the database
• REQUESTS list of all interactive model requests on Swiss Modeling with project urls and status
• PAGES switch to any page currently available. Empty pages will be removed in hour or so after their creation.