Take the 2-minute tour ×
Biology Stack Exchange is a question and answer site for biology researchers, academics, and students. It's 100% free, no registration required.

After HGP, we are not having many databases which consist of several notepad files of ATCG....

Can we distinguish quantitatively a given A,T,C and G stretch as DNA or Gene?

share|improve this question
    
What exactly do you mean with your first sentence? There are many databases containing nucleotide sequence data. One of the biggest is GenBank, which could be considered a collection of "notepad files of ATCG". Also, except for RNA viruses, all genes are encoded in DNA. So it does not make sense to say "DNA or Gene". It sounds like you are asking how gene prediction works? –  jarlemag Mar 3 at 9:27

3 Answers 3

up vote 4 down vote accepted

I interpret your question as: Given a stretch of DNA sequence, can we determine if it encodes a gene? My summary of the answer would be: "Sometimes".

The problem you ask about is called "Gene prediction" and is described in some detail by Wikipedia: http://en.wikipedia.org/wiki/Gene_prediction

The most basic indication that a gene may exist is the presence of an "Open Reading Frame" (ORF). An ORF is a stretch of DNA from which a single protein may be produced. However, an ORF is not strictly required for a gene (as not all genes encode proteins), and not all ORFs are genes. There are many software tools for gene prediction available which use different rules to predict if a given region of DNA is functional (contains a gene or a regulatory element). They may be usable on any kind of DNA sequence, or on sequences from certain organisms only, depending on the program.

Two commmonly used gene prediction tool are GLIMMER (http://www.ncbi.nlm.nih.gov/genomes/MICROBES/glimmer_3.cgi, used for microbial genomes) and GeneMark (http://opal.biology.gatech.edu/gmhmm2_prok.cgi). See also http://en.wikipedia.org/wiki/Gene_prediction

As a quick check, you may also use NCBI BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) to see if your sequence aligns to a known gene.

share|improve this answer
    
Thanks a lot for making me understood about the principle. I am anxious really how the genes are ordered as a finite length string of four bases? How random as a finite string of four bases when we consider it as a gene? My conviction is that genes are not randomly arranged as a finite string. There must be a beautiful organization. I would like to know in detail about the organization of them. And definitely that organization would be missing in any arbitrary DNA stretch. –  Sk Sarif Hassan Mar 3 at 10:36
    
Yes, the statistical properties of coding and non-coding sequences are different, and this is exploited by some gene-prediction programs. Note that some non-coding sequences might be the remains of a gene which is no longer expressed, and so may have similar statistical properties as an expressed coding sequence. Such sequences are called "pseudogenes". The following links might be a place to start: NCBI Book on computational genomics (ncbi.nlm.nih.gov/books/NBK20261) and a review on identifcation of protein-coding genes: genomebiology.com/content/10/1/201 –  jarlemag Mar 3 at 10:54
    
Thanks a lot, Man! –  Sk Sarif Hassan Mar 3 at 11:17

This is possible and is an important topic in bioinformatics. Lots of tools have been written and papers have been published, most important

  • A random sequence usually "encodes" only very short protein sequences before the "stop" codon (three nucleotide combination marking the end of the protein) is met just by chance. Coding sequence encodes a reasonably long protein.
  • Same amino acid can usually be encoded by multiple alternative sequences of three nucleotides (codons). In a real coding sequence they are not equally probable, some codons are preferred over others (which ones, is organism-specific).
  • It is possible to find characteristic nucleotide patterns inside or in some cases near the coding sequence. Dropping the coding sequence alone into genome would only result in a non-functional pseudogene.
  • Most often, encoded protein sequence is more or less similar to the sequence of some already known protein.

.. and so on. Try GenScan, for instance.

share|improve this answer

My conviction is that genes are not randomly arranged as a finite string. There must be a beautiful organization.

In bacteria, you will have a whole series of genes required for a pathway all next to each other, so that their expression can all be controlled from one point upstream of them. We call that whole suite of genes and the upstream binding site an operon.

But in eukaryotes, there is no grand organization of genes with regard to their position on the chromosomes, in general. Regardless of what you think they should look like, we don't observe a wonderful organization. Evolution results in things looking pretty haphazard. Go browse ensembl.org if you want to see for yourself.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.