HG19 Unlocalized Contigs

[Bio.X2Y]

Hi,

I was wondering if someone might be able to shed some light on the "validity" of the unlocalized contigs that are included as part of the UCSC HG19 assembly (not the haplotypes).

In particular, do these contigs reflect:
(a) chromosome segments that are definitely known to exist, and are not already present on the traditional chromosomes, and will in time get localised.
(b) segments that may or may not exist (e.g. their apparent existence might be an artifact of the assembly process), and so may never be localised.

I'm asking because I'm trying to get a feeling for whether it is appropriate to include these in an rna-seq mapping exercise.

Thanks!

[NextGenSeq]

I don't think anyone knows for sure but I vote for a

Also, some are mitochondrial DNA.

[vinay052003]

By definition these contigs are "yet to be placed" fragments of the chromosomes. These contigs possibly arise from highly repetitive regions such as centromere or other highly variable places in the genome (due to transposable elements and sites of high recombination rate). These contigs have gene annotation tracks and other kinds of annotations similar to well mapped chromosomal sequences. So I guess there is no harm in including these sequences while mapping sequencing reads.

[Brugger]

Just to clarify things: There are two different classes of non-placed clones in GRCh37 (the official name, not hg19, as ucsc calls it).

Unplaced: with means that is it known what chromosome a clone belongs to, but cannot be placed in the TPF with confidence. The mapping to chromosomes was done in various ways including FISH.

Unlocalized: which means that they belong to the human genome, but to what chromosome they belong is yet to be established. These include clones with repeated regions that can go to multiple chromosomes. I seem to recall that the chr22-p clones went there because the region is so repetitive.

So to answer your question, the clones are placed in these two classes until they can be placed correctly in the TPF and included in the genome. The goal of the GRC (Genome Reference Consortium), is improve the human assembly now that the sequencing technology has improved.

As far as I know (I used to be part of GRC) there is no mitochondrial DNA in the unplaced/unlocalized clones, if you find any please report this to the GRC: http://www.ncbi.nlm.nih.gov/projects.../contact.shtml

[freeseek]

HG19 Unlocalized Contigs

Brugger, could you shed some light on why some contigs are unplaced and some are unlocalized? Where they all analyzed using FISH? And if so, why most of them were not placed? Is there any reference that correctly explains how these contigs were handled?