discrepancies between cDNA and protein sequences

[NICHOLAS THEODORAKIS]

In article <3aqa0d$ad2 at news.univ-rennes1.fr>,
Le Pennec Jean-Paul <Lepennec at univ-rennes1.fr> wrote:
>I have analysed a cDNA sequence of a protein already sequenced by 
>chemical-physical analysis and I get differences of the amino acids 
>composition. I'd like  to know wether i can trust 100 % the protein 
>analysis or go back to my sequence.What do you think? I am looking 
>forward hearing from you !

That depends.  Look at the two types of sequences and think what kind of 
errors would give you one sequence vs. the other.  For example, in the 
protein seq., is there a Glu that could really be a deamidated Gln?  If you 
look at the primary data for the protein seq., if available (by Edman 
degradation?), what is the confidence of the identification of a given 
residue, especially if it's low yield?  Is there corroborating evidence 
for the protein sequence, e.g., comparing Edman vs. mass-spec.?  
Alternatively, consider the DNA sequence.  Look at the primary data and 
see if the contested region has really been overlapped by two different 
clones and in both orientations.  Is it in region of the gel that is 
difficult to resolve?  If so, can an alternate reading of the sequence 
give the sequence from the protein data?  Lastly, also consider that there 
might be polymorphisms.
-- 
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
			Nick Theodorakis
			ntheo at welchlink.welch.jhu.edu
			Johns Hopkins Medical School, Baltimore, MD