Frequently Asked Questions
**If you have specific questions that were not addressed in the about page, help pages or here, feel free to write us and we will add them (firstname.lastname@example.org).
How much time takes a PROSS calculation?
The time required for PROSS to complete a calculation depends on the size of your protein and on the number of other queries running in parallel. You may receive the output email within few hours or within 1-2 days. If after 2 days you haven't received the output email, checkout the next question and if the answer is not there please contact us (email@example.com).
Can I submit a structure with missing density (missing residues/loops)?
Yes! You may submit a structure that has some missing residues.
Obviously, the missing residues will not be mutated, PROSS is not aware to their existence. In addition, 3 residues from each side (N and C) of the missing part will not be mutated either (because we can't be sure what is going on there from a structural point of view). In the final output however, you should get the stabilized sequences with the missing residues, but we strongly recommend that you align the final sequences with the original sequence you typically use to make sure that there wasn't any problem with the addition of the missing residues (the addition to the final sequence is based on the fasta file downloaded from RCSB or provided by the user in case of using the upload structure option).
Also note that in case you use another structure/model to define specific residue numbers to fix, make sure that the defined residues are not missing in the structure that you actually submit. Do not refer to residue numbers that are not in the submitted structure.
Can I submit a PSE file created by Pymol?
No. You can't submit a PSE file. The format of submitted structure should always be pdb.
In Pymol there is an option called "save molecule" that allows you to save specific molecules in a pdb format.
Can I submit a NMR structure?
No, you can't.
Though, if you tweak the NMR-based pdb file to look as if it is based on a crystal structure (having all atoms appearing only once), you can submit. However, we strongly recommend avoiding NMR structures, they are typically not accurate enough for these kind of calculations.
Can PROSS handle non native amino acids (e.g. Selenomethionine MSE, carboxylated Lysine KCX etc.)?
PROSS turns Selenomethionines (MSE) to Methionines, Carboxlyated Lysines (KCX) to Lysines and Hydroxycesteines (CSO) to Cysteins. Other non native amino acids are recognized and spliced out. This does not affect calculations but mutations at the protein regions near the spliced amino acid (<5A) are less reliable. Consider removing them from sequences that you plan to order.
If you specify position numbers of non native amino acids that are spliced out you will get an error.
Is PROSS good for stabilizing symmetric oligomers?
No and yes. In PROSS you can design only a single chain per query. If your protein is a symmetric dimer with chains A and B for instance, you must select either chain A or B for design. If chain A was selected, chain B will be removed and PROSS "will see" a solvent at the interface region. We therefore recommend to avoid any design of chain A residues at the contact site between chains A and B. This can be easily done using the interacting chains option.
In some cases (especially for higher order oligomers) using the interacting chains option will fix too many residues leaving PROSS with too little to work with. In such a case we recommend that you apply your own decisions using the manual fixation option specific residues to fix
Is PROSS good for stabilizing antibodies?
Yes, but results are probably less reliable than for other families. Specifically, mutations at CDR regions are not reliable.
If you wish to stabilize an antibody for which there isn't an available crystal structure, contact us. We may be interested in that.
Is PROSS good for stabilizing membrane proteins?
For modeling PROSS uses a standard energy function designed for soluble proteins. It does not contain any energy terms that take into account the membrane environment and in practice, it is unaware to the membrane and its location.
Is PROSS good for stabilizing small proteins (<100 amino acids)?
PROSS follows a conservative approach in order to minimize false positives. In very small proteins, you might end up with no mutations at all or with a very small number (<4) that is less likely to produce an effect. This is especially true when there are a lot of constraints (small-molecule ligands, interacting chains, specific residues were fixed). We suggest to submit free-of-constraints queries followed by removal of mutations that seem too close to active sites by eye
I submitted a query and received an error email. What should I do?
Errors are most likely due to user's wrong input.
In the error email you have a list of all the parameters that you submitted to PROSS. Go over each parameter and make sure it is correct.
Specifically, users who use the upload_pdb option and/or the upload MSA option tend to submit files that do not obey the format rules as described in the help pages.
Here is a list of common user errors that we saw in the past, listed by categories:
- General errors
- The designed chain has more than ~820 amino acids. If this is the case, the job might require too much memory and will therefore fail. You are welcome to contact us to verify if this was indeed the case (firstname.lastname@example.org)
If you have more ideas about errors to include here let us know.
- Errors in PDB id (if you uploaded your own PDB file see a section below)
- The desired PDB id contains the letter 'o' but you typed the digit 0 or vice versa.
- Wrong id - copy the id reported in the email and paste it in the rcsb web. Make sure it is your protein of interest.
- The id is of a NMR structure - PROSS is not compatible with NMR structures.
- The PDB files contain negative residue numbers. PROSS is incompatible with negative numbers.
- (rare) The PDB file contains a residue for which one or more of the backbone atoms is missing. Note that if an entire residue is missing this is not a problem. However, if some atoms exist, then all backbone atoms should also exist. To solve this problem remove all other atom lines related to the specific residue (i.e., turn it into a missing density).
- Errors in Chain identifier to design
- Wrong identifier - not of a protein chain but of DNA or RNA chains.
- Wrong identifier - the identifier is of a protein chain but the wrong one and is incompatible with the rest of the parameters.
- Numerical identifier - PROSS is incompatible with numerical identifiers (1, 2...). You can change the identifiers to letters and resubmit the query using the upload files option.
- Errors in Small-molecule ligands
- Wrong identifier - either metathesis or maybe you are using a version of the name from the literature. Make sure you are using the exact name used in the pdb file.
- Special ligand - for instance, the ligand is a free glutamate with the identifier 'E'. This is also an amino acid in proteins and PROSS does not know how to discriminate between the protein amino acids and the ligand. Edit the pdb file and change the ligand name; then use the upload files option to resubmit the query with the edited file.
- Errors in Interacting chains
- One of the identifiers was already used in the Chain identifier to design option.
- Errors in Specific residues to fix
Very common(!) and happen when one or more of the numbers does not exist in the pdb file. Common reasons:
We suggest to open the pdb file on PyMOL; then go over all the position numbers you entered here and make sure you can select them on the deisigned chain in PyMOL.
- A missing comma leading to the merge of 2 numbers.
- Using a numbering scale from a paper or from a database that does not match the numbering scale in the pdb file.
- Trying to fix a position for which there is missing X-ray density.
- Specifying a number of a non native amino acid that is not MSE/KCX/SCO (see here).
- Errors in Upload your own multiple sequence alignment
The file should obey a strict format and we suggest that you read the MSA help page.
The most common error is that the query sequence in the MSA does not perfectly match the FASTA sequence file: if you uploaded your own PDB file, then you also had to upload a FASTA sequence file. If you used the PDB id option, PROSS downloaded the FASTA sequence file from the rcsb web. Compare your query sequence in the MSA to the sequence in the FASTA sequence file. Make sure that they are identical (obviously the query in the MSA will have a lot of hyphen characters due to the alignment with all other hits, but after removing these hyphens it should be identical to the FASTA sequence file)
- Errors in Upload your own PDB file and a Sequence file
- A PROSS bug
If in your error email the PDB id or the Sequence file fields are empty - this is in fact our bug and we are working to solve it. It is probably related to connection issues so try to wait few hours and then submit again using the same files and parameters.
- User errors
Errors in these options are common either due to wrong format of one of the files or due to incompatibility between the 2 files.
- You uploaded an NMR solved PDB file or a PSE file (generated by PyMOL). These formats are not allowed.
- The sequence titles in the FASTA sequence file are wrong - for instance, you forgot the : sign between the name and the chain identifier.
- The PDB file has amino acids that do not appear in the sequence file. For instance, the PDF file has a HIS-tag at the end that the FASTA sequence file does not show. Remember, the sequence file should be equal or longer than the sequence in the PDB file.
- The PDB sequence and sequence have point mutations compared to one another. This is not allowed.
- You also uploaded a MSA file and the sequence file and the MSA or incompatible with each other. Read the help pages.
- Your uploaded PDB file must have the suffix .pdb and nothing else after.
- One of the uploaded files has a character/sign that are not allowed. For instance, a whitespace before an ATOM line in the PDB file.
- One of the uploaded files has a hidden character that is not allowed. This can happen if you prepared the file in a rich text editor and is even more likely if you used a PC (windows) to prepare the files. Try to open your files in other editors or other operating systems (mac) to check if this is the case.
- See also the common errors in the PDB id option - your own files might suffer from some of them.
If you think that all the parameters are correct and yet the job fails, please contact us at email@example.com.