Biocomputing

Gibson Group

EMBL

The aim of this practical is to get hands-on experience running some Web Servers for eukaryotic gene prediction on an "unkown" sequence. We will use coding prediction programs together with promoter, splice and poly-A site prediction programs. All the information gathered has to be combined to find the coding regions of the gene. There is not time to go into the theory behind the prediction methods.

As anyone who has looked at automatically annotated genome sequence data will know, automatic gene prediction in higher eukaryotes does not work very well. Perhaps this is not surprising, considering that differential and tissue-specific splicing, genes within genes, TATA-less promoters, the rare class of U12-dependent (AT-AC) splice sites, the need for species-specific parameters, all complicate the problem.

In practice, the bench biologist is best advised to inform themselves of the species-specific signals (e.g. mammals tend to have a looser splice site consensus than the worm) that they should expect and then run several servers and synthesise the results, looking (in part) for consistency between different servers. As there is a lot of activity in gene prediction (even though much of it is "reinventing the wheel") we may hope for better predictions and better presentation in the future.

Several web sites maintain useful lists of gene prediction services e.g.

 Gene Prediction Links of the Bork Group
 CNRS Marseille
 Rockefeller University
 Atelier Bioinfomatique, Universite Aix-Marseille

WWW Servers used in the Gene Prediction practical

We will use:

• NetGene2 and HMMgene to predict coding exons and splice sites in eukaryotes.
• The LBL Promoter prediction service and TSSG, TSSW from the Sanger Centre.
• The POLYAH server for the prediction of the poly-A site.
• GeneWise (from the Wise2 package) for aligning protein v. spliced DNA sequence.

Step 1  Getting the DNA sequence

For the practical we have choosen a human genomic sequence that may perhaps encode an adenylate kinase...

• Open a new navigator window and load this page into it.
• Click here to get the ~12 kb sequence (press on the middle mouse button to open a new browser window).

Step 2 Looking for a Promoter candidate

Promoter prediction is heavily dependent on finding good matches to the TATAAAA motif. Further clues may be provided by CCAAT-Box, CpG islands and other transcription factor binding sites. But if you run MatInspector with the TRANSFAC DB on our query, you will be astonished at the profusion of candidate sites throughout the sequence - so we won't bother. We will run two promoter prediction programs and tentatively assume that the best intersecting promoter is the correct one.

• Open a new navigator window and load this page into it.
• Load the LBL Promoter query submission page.
• It is worth familiarising yourself with the layout and note the on-line help.
• Cut and paste the query sequence into the sequence box and submit the job.
• When the result arrives look at the set of predicted promoters
• Can you see matches to the TATA box consensus tATA^A/_TA^A/_T
• Which promoter has the silliest TATA-box?
• Open a new navigator window and load this page into it.
• Load the TSSG query submission page.
• Toggle on the TSSG button.
• Cut and paste the query sequence into the sequence box and submit the job.
• When the result arrives look at the set of predicted promoters
• How many TATA boxes were found?
• Are the listed transcription factor binding sites informative?
• How well do the two searches agree on candidate promoters?
• How many candidates do they both find?
• Is there a single best candidate from combining the searches?

In mammalian genes, polyadenylation sites are usually preceded by AATAAA or ATTAAA ~20 bases before the cleavage site and followed by a more weakly conserved GT-based motif. While these motifs are trivial to find, they only function in the right context - which is harder to define and includes regulation by upstream splicing factors. An important rule to remember is that there should not be an in-frame stop codon in an internal exon ie the true translation termination will be in the last exon. (Violations to this rule suppress mRNA production, to the cost of many experimentalists, and is occasionally used for differential mRNA regulation in vivo e.g. for certain Ig splice variants.)

• (As needed, open a new navigator window and load this page into it.)
• Load the POLYAH query submission page.
• Toggle on the POLYAH button.
• Look at the POLYAH help and note the quoted prediction accuracy.
• Cut and paste the query sequence into the sequence box and submit the job.
• When the result arrives, look at the predicted poly(A) sites.
• How many candidate sites were found?(You need to open the numbered sequence to see this.)
• If one or more of these sites are false is the prediction accuracy as good as claimed?
• How might overprediction of poly(A) sites be avoided?

There are a number of servers that separately predict splice sites and coding sequence bias but this information needs to be analysed together. We found that the CBS site in Denmark could provide all the information, though from two different servers. The NetGene2 server provides a graphical postscript output that we can print out and mark our predictions on. From the same group, the HMMgene server (using different algorithms) provides list output including potential Start and Stop codons. Both servers overpredict splice site candidates. In case you need reminding, classical splice sites look something like:

• Donor Consensus:                              ^c/_aAG^GT^A/_gAGt

 
• Acceptor Consensus:                     (T>C)_nN(C>T)AG^gt

• (As needed, open a new navigator window and load this page into it.)
• Load the NetGene2 query submission page.
• Paste in the sequence and submit the job, which takes a few minutes to run.
• The output provides a list of candidate splice sites (on both strands) and a graphical coding/splicing prediction.
• However it is not clear which translation frame is supposed to be coding.
• It is worth printing this figure out and using it to summarise our prediction attempts!
• Click on the Direct strand link and save the compressed postscript output (has a .gz suffix).
• Open a UNIX X-window (terminal from the desktop)
• Uncompress the file by typing UNIX command
• gunzip filename.ps.gz
• Print the file to the printer outside room V111 by typing
• lpr -Plj-v111 filename.ps
• Now load the HMMgene query submission page.
• Get the FASTA format sequence and paste in the local file box.
• (These servers are rather fussy about sequence formats!)
• select 5 best predictions and toggle on predict signals.
• Submit the job.
• The output appears confusing:
  • It is a list of predicted exons, coding spans, coding starts and stops and splice donors and acceptors.
  • Click on the Explanation link to understand the output format.
• We can now begin to assemble a complete gene prediction.

We now have predictions for all the components needed to assemble the gene, rather inconveniently spread over many separate web outputs. We have to manually assemble all this into one prediction. This can be done on the Netgene2 and DNA sequence outputs using biro and fluorescent marker. The following guidelines may help.

• Start from a strong point such as:
• A well-predicted internal coding exon with good splice borders.
• Work forwards and backwards toward the promoter and poly(A) boundary signals.
• Reported splice site quality is not a completely robust guide to usage.
• Context dependence is also important.
• Splice sites tend to be overpredicted.
• Some (true) splice sites might be better predicted by the HMMgene algorithm than by NetGene2...
• The terminal exon should be partially coding, including the stop codon and the poly(A) signature.
• The initiation codon should obey the Kozak rules:
• It is normally the first methionine from the 5' end of the mRNA.
• At least one of the underlined residues should be present in the consensus APuXXAUGG.
• Once the prediction is completed, we can check it in the next exercise.
• Good luck!

Usually nowadays, related sequences are already present in the databases. When available these may be the fastest way to get a good gene prediction. Often this prediction will be more reliable than the coding bias predictions though one should be aware of the possibility of sequence error, differential splicing etc. and of course finding the coding exons is not a complete gene prediction. The GeneWise program has an exhaustive (slow) algorithm to align a protein to a DNA sequence, allowing for splice site recognition. (In a real situation, BLAST programs would be useful for first picking up the matches in a DB search.)

• Open a new X-window and type rlogin tau. Give your username and password.
• Type prepare wise2 in the window. (prepare is a way to set up program environments on EMBL UNIX).
• We've prepared files with the human DNA and a homologous chicken protein to compare.
• Now you can type or cut and paste the following command into the UNIX window:
• genewise /home/seqanal/public_html/courses/Jul01/kad1_chick /home/seqanal/public_html/courses/Jul01/hsak1.dna
• GeneWise will run with default parameters and after a couple of minutes will print out the matched exons.
• Now compare the results to the predictions so far
• How many Exons are found?
• Are the splice sites between or within codons?
• Did you find all these coding regions earlier?
• Have we now found all the coding exons (the chicken homologue has 194 AA)?

 
• Now lets look at the annotated genomic sequence entry for our test sequence, HSAK1
• Note that no cDNA has been sequenced for this gene: gene structure was inferred by some transcript mapping and by protein homology.
• Most of the elements of the gene are listed in the feature table.
• Did you get the promoter?
• Did you get the starting methionine? Does it obey Kozak's rules?
• How many amino acids are in the first coding exon?
• If you made any errors in the prediction, can you see where you went wrong?
• There is a problem with the annotation of the first intron's acceptor:
• do you think this is -
• a). an unusual splice site?
• b). an annotation error made by the authors?

There seems to be no single high quality tool for doing eukaryotic gene prediction work. The variations in the results from prediction servers indicate that there is scope for improving the algorithms. Graphical presentation of the results is patchy - for example we need to know start, stop codons and which frame has the coding potential, information that we did not get from the graphical plot. To do this for real, we would need to assemble results from many servers and work with a hard copy of the DNA sequence and it would take longer than the morning we set aside today, to do the job properly. In fact the Staden package has for many years been able to produce a plot with all this information, although it was clumsy and old fashioned to use and some of the prediction methods may be more sensitive now. This package has been updated recently and we will have a look at how useful it is for gene prediction. 

You can find this page at http://www.embl-heidelberg.de/~seqanal/courses/Jul01/GenePred.html