Nyhetsflöde

Please note the updated deadlines. Sorry for the confusion.

Hi Lars! I was wondering if it's possible to have an extra lab exercise on Friday afternoon or similar, as many of the Bio-students have a mandatory class with presentations at the same time on Friday. Mainly because that will be the last opportunity to present labs/get help with the project. 

I will look into it. I am not available at another time, unfortunately.

When can we expect result from the project?

This week. I am handling a few every day.

How many members (minimum and maximum) are expected to be in a group for the project ? 

Groups of two are preferred, but 1 to 3 members are accepted.

How do you want us to hand in the report and notebooks? Email?

Yes, email is good, containing PDF or link to something readable. I do not want files in formats like Word, OpenOffice, etc. 

When are we going to receive an evaluation for the project report?

Next week.

Any chance  the projects be marked by tomorrow?

Hi!
The links seem to be broken. Also, how many domains should be looked at and how should they be choosed?

Both links have now been fixed. I hope this was not a big concern!

The goal is to study all human genes to see what domains they contain and whether they cross exon boundaries or not.

Question about the definition of noisy columns:

•  at least 50% of amino acids are unique

Does this mean that for example the column "AFFFFFFFF" would be considered noisy because there are only two kinds of amino acids and 50% are unique? Isn't that a good alignment?

How do we use fastprot and fnj on the CSC computers?

Edvin, I interpret the statement in the first question as "a column is noisy if at least 50% of the entries in the column are unique". In "AFFFFFF" only A is unique, as there are multiple F-entries in the column. Would that make sense?

Nikos has the interpretation I want. I added some simple examples to help interpretation. 

I have now added instructions for how to get access to FastPhylo programs. The solution is to run two 'module' commands that set up your path.

Ok I understand. Thank you!

I am unsure of how to load the data into my database form python. In other words, I don' t know how to write this line: sqlite> .read /afs/nada.kth.se/info/appbio10/data/protdb.sqlite3 using Python.

The 'connect' function in Python's sqlite module will work if you have created the database with something like:

sqlite my_db <  /afs/nada.kth.se/info/appbio10/data/protdb.sqlite3

Hope this helps!

Thanks

Just to make sure I get this right: we are not only interested in the presence of a signal peptide (i.e. there or not) but in the actual sequence of the signal peptides we find?

No, it is the presence that is the main question. Once that decision is taken, retrieving the signal peptide sequence is fairly trivial task. Is it "Problem 3" that confuses you?

Are we free to use whatever software we like, or should we use Python?

You must use Python.

We get a 403 error when trying to access the database by importing qblast on the school computers but not on our personal. However, everything works when we copy the NCBIWWW source code into our script.

Has anyone else had the 403 error and managed to solve it?

It might be that the installed version of BioPython is too old. NCBI may have changed the underlying protocol. Has anyone successfully managed to run this on a school computer?

Even though I can access alignment.hit_def and alignment.hit_id, I'm having trouble finding the Hit_accession. Is it the same as alignment.accession? It does not seem so, as it is a number, while the example output from test2.xml is CYTC_MOUSE. Is the example output correct?   

That number is also an accession, but use the definition line instead.

To avoid others having a bad time: temporary files created by tempfile can't be opened by other programs in Windows.

Hey, we don't understand what the "Phylip bootstrap" analysis is. What is the command line argument to run this and where can we find information about how it works?

http://evolution.genetics.washington.edu/phylip/progs.algs.data.html

This is linked, but which one of them is the bootstrap analysis?

As I understand it, you should run several phylip programs in sequence. - phylip seqboot: bootstrapping your samples to generate more (say K samples) - phylip protdist: computing the corresponding K distance matrices - phylip neighbor: computing the K newick trees - phylip consense: merging K trees into one. Now exactly how to do all these steps from within Python I do not know yet.

The seqboot documentation has useful information on performing bootstrap analyses using phylip. Unfortunately the entire process is rather painful, but quite doable.