In this lab, we'll charcterize proteins using computational tools. There are two proteins at the top of the list of links, we'll be characterizing them. The first protein is from Gram-negative bacteria B. pseudomallei and the second protein is a one of the myosins found in dog (C. familiaris).
- 1. Use the ProtParam program linked from the ExPASy page and use it to calculate the MW and pI of the two proteins. Remember to remove the N-terminal methionine. ProtParam is picky about sequence format--check that the protein sequence you enter is what ProtParam sees.
- 2. Take the two proteins, and search for known domains using InterPro. For each protein give the InterPro domains or families and a short description. InterPro will list multiple family and domain entries for each protein domain. InterPro draws annotations from numerous sources databases and each database adds to the description of a domain. Proteins may be annotated with one or several InterPro domains. In your answer provide a single annoation for each distinct domain in the protein. Don't include multiple domains that annotate the same region of the protein and are redundant.
- 3a. Now find whether these proteins are predicted to have membrane spanning segments. For each protein, run TopPred, examine the predictions and give the number of predicted membrane spanning segments, and predicted intra/extracellular location of the N and C terminal regions. Select the 'Produce an image for each topology' setting. Also note that TopPred has a setting for prokaryotic/eukaryotic sequence.
- 3b. What is the default hydrophobicity scale used by TopPred?
- 3c. In the topology plots KR values are shown. What are they and why are they shown?
- 4a. Now use the Phobius server to predict transmembrane segments and the topology of the two proteins. Are the same number of transmembrane segments predicted in the same places by both programs, and is the predicted topology the same? Answer for each protein separately.
- 4b. For the first transmembrane segment of 1P7B_A, what aa's are predicted to be found in this segment by the two programs? Examine the differences: is the length of the predicted segment different, or is it shifted, or both?
- 4c. Does Phobius predict both membrane spanning alpha helices and beta strands?
- 4d. Give the citation for Phobius.
- 5a. Now lets try to predict the cellular location of these two proteins using PSORTII and TargetP. Run TargetP on the C. familiaris protein and PSORT on each protein using the apporpriate bacterial or eukaryotic PSORT program. Give the predicted cellular location of the proteins. It is interesting to follow the links in the PSORT WoLF results "details"->"PSORT features and traditional PSORTII prediction" links to see the details of the prediction.
- 5b. Describe in a sentence or two how TargetP makes its predictions.
- 5c. Briefly describe how the PSORT II prediction approach differs from that of TargetP--what is the biggest difference between them?
BIO520
Site maintained by Jim Lund