[Genome] Not the same assemblies in the browser and in the downloadable sequence files ?

[Herve Seitz]

Dear developers,

I noticed a difference between the genomic sequences for Drosophila  
melanogaster chromosome X on the browser and in the sequence file  
that I downloaded from http://hgdownload.cse.ucsc.edu/goldenPath/dm2/ 
bigZips/ :
according to the browser, the first 30 nucleotides of melanogaster  
chromosome X are: TAGAGCAAAAAATAGACATTTTAATGGCGC (cf http:// 
genome.ucsc.edu/cgi-bin/hgTracks? 
hgsid=87534567&hgt.dummyEnterButton.x=0&hgt.dummyEnterButton.y=0&clade=i 
nsect&org=D.+melanogaster&db=dm2&position=chrX% 
3A1-30&pix=620&hgsid=87534567 )
but according to chrX.fa (in the archive chromFa.zip, downloaded from  
http://hgdownload.cse.ucsc.edu/goldenPath/dm2/bigZips/ ), the first  
30 nucleotides of that chromosome are: CAACATTAGCGCCATGCCCACTGTGGGGAA.

When I search this sequence in the "browser" version of the  
chromosome (with BLAT), I can see that it starts on bp 6547 of  
chromosome X; reciprocally, when I look for  
TAGAGCAAAAAATAGACATTTTAATGGCGC in chrX.fa, I find it on the 164th  
line of the file (hence, at position ~8100).

So it seems that the downloadable sequence for D. melanogaster  
chromosome X, and the browser version, do not result from the same  
assembly. Is is possible to download the "browser" version of the  
assembly ? It seems that chromosome X is not the only one - chr2R  
also differs between the downloadable file and the browser version ;  
the beginning of chr2L seems to be identical in both versions, but  
the total length is different: 22,407,834 bp according to the  
browser, and 23,471,775 in chr2L.fa.

Thanks in advance,

		Hervé

--
Hervé Seitz, PhD
Department of Biochemistry and Molecular Pharmacology
University of Massachusetts Medical School
Lazare Research Building, Room 870P
364 Plantation Street
Worcester, MA 01605-2324
USA
telephone: (508) 856 1220