COT 6936/CAP 6990 Bioinformatics: Fall 02
Frequently Asked Questions

Oct 13:

Question: 
    In part A, the human U1A with gi:2554638 has only one domain, 
    while all other sequences in the problem have 2 domains. Can you 
    tell me which one I should use for the human sequence?
Answer: 
    Yes, gi:2554638 is not the U1A protein, but only the 2nd domain.
    You should be using sequence with gi:134092. 

---

Question: 
    In part B, how can I calculate the "log odds scores" for A and T in
    column 1 since that column only has C and G?
Answer: 
    Since the emission probability for A and T is 0 for that column,
    the log-odds score should be negative infinity.

---

Sep 29:

Question: 
    I have trouble opening the SWISS-PROT page:
    http://www.expasy.ch/sprot/
Answer: 
    Instead, try the mirror page at http://us.expasy.org/sprot/

---

Question: 
    I can't open the link for gi|3041867 (in the list of the 5 
    search results). I get: "Detected cycle in the query list for #1"
Answer: 
    See the answer for the next question. 

---

Question: 
    I cannot get the "Graphics" display option to work for p53.
    What should I do?
Answer: 
    There seems to be a problem with that display option for p53.
    Hopefully, it is only a temporary problem with the NCBI page.
    GenBank does have temporary problems infrequently. After all,
    thousands of people use it everyday. 

    For now, parts of questions 3, 4, and 5 that deal with specific
    exons that you cannot answer, you may ignore. We will revisit this
    later. Meanwhile, I suggest you try protein p73 instead to play 
    with this option.
    
    If you search for p73 using the query "p73[Protein Name] AND 
    Human[Organism]" in the Nucleotide database, then you get many 
    entries. Most of them are for specific exons. But one of them 
    is for the complete chromosome on which p73 lies. Navigate and
    see if you can repeat the process for this protein. 

---

Sep 28:

Question: 
    I found the search could not get anything if I searched with  
    "p53[Protein Name] AND Human[Organism]". Then if I changed the 
    query to "p53[Protein Name] AND Human", I could get 6 records, 
    which was almost the same as what you described. 
Answer: 
    It is true that the database is not behaving the same way that
    it did a few days ago. They must have either updated their database
    or upgraded their software. However, if you present the first query
    "p53[Protein Name] AND Human[Organism]" to the "Nucleotide" database
    instead of the "Protein" database, it works fine. 

---