ProteinCenter User Manual
Table of Contents

Chapter 27. FAQ

An FAQ is available at Proxeon's website as well.

Q: Why is the description "unknown" when there is a Gene Symbol for a protein?

A: The description shown depends on the preferred protein key. The system is always forced to display a description. In cases where multiple protein keys exists for a preferred type of ProteinID. The system is optimized to display the most optimal description, however in certain cases none of the descriptions are optimal (e.g. if only a few old GenBank records e.g. because the protein sequence is a fragment). Nonetheless the sequence may be mapped to a gene as it may be a close allele or fragment from that gene. The protein description is the original description, and hence the description "unknown" may very well be outdated as the gene and other isoforms of the proteins are annotated over time, but there exists no other descriptions for that particular protein sequence.

Q: Why can't I import a list of genes?

A: ProteinCenter is a protein-centric system, so the gene entries will be converted to protein entries upon import. A gene extension allowing import of genes is being considered. However, for NCBI protein keys you can upload a list of nucleotide records to Entrez using "Batch Entrez" (part of Entrez Tools) and convert the entries to proteins using "Display Protein Links" found in the display drop down menu.

Q: Why can't the import function read my input files?

A: The import function requires that only the columns described in Chapter 21, Import are included, and that no other columns are included. The file will not be imported, if it contains extra columns or columns with a wrong header.

Q: Why aren't my protein keys the same as the ones in the imported files?

A: ProteinCenter shows the most optimal protein keys for a preferred type of protein keys. Therefore, it may choose another than the one in the import file. This ensures a consistent data analysis. The user can always see the imported protein key by looking at the protein data view.

Q: I have imported a dataset of a given type of accessions (e.g. IPI) - when I export it some of the IPI columns are empty - even though all the IPI accessions was known by the system - and they are found in the KEY column in the exported data.

A: The reason is that only live accessions are exported, so the missing entries are those that are outdated (hence a valid accession from this type of accession does not exist anymore).

Q: I am trying to make a compare folder - but I cannot select more than one folder at a time

A: You have probably selected the folder by clicking the folder name (that then get a blue background) - instead you should click the yellow folder icon and a green check mark will appear - then you can select other folders the same way. Deselection of selected folders is accomblished by clicking the folder icon again.

Q: Why is the user interface all messed up after updating the ProteinCenter application?

A: Browsers usually cache style sheets to improve performance. If changes are made to a style sheet a browser won't always notice and just keep using the old one. This may cause layout artifacts on a web page. To force an update of the style sheet in Internet Explorer or Mozilla FireFox, use Ctrl + F5.

Q: Why are so many of my proteins lacking a description?

A: Chances are that the preferred protein key your ProteinCenter is set to Ensembl. Ensembl does not provide descriptions with their protein records. However, it is likely than one of the other protein sources i ProteinCenter (GI, IPI or UniProt) have a description for the protein. The Keys section of the ProteinCard shows description from all protein sources.

Q: How can I search for a certain ProteinID within my dataset?

A: The lookup performs a global search on all the Protein identifiers in the ProteinCenter database. To do a search within your dataset use the accession key filtering functionality described in Section 11.3.1, “Accession Key”

Q: When I import a file containing protein records the resulting dataset in ProteinCenter is empty (i.e. contains no proteins).

A: The most common reason for this that the input file is not of the selected import format. By checking the Import Format tag on μLIMS page for the empty data folder, you can see the import format used by the import procedure.

Q: I would like to use the graphics shown in statistics, but the resolution is not exelent when doing a screen dump?

A: You can click on the image as described in Section 25.5, “How to copy images from ProteinCenter”. Right click on the larger image and select 'Copy'. In the target application select 'Paste Special' and select 'Device Independed Bitmap'.'

Q: What are the rules for mapping peptide sequences to protein sequences?

A: Generally a peptide is mapped to a protein if the peptide's amino acid sequence is a substring of the protein's sequence. But there are two additional rules: