how are physical intervals calculated in the anchoring process?

[Pietro]

Dear support group,

I would like to ask what's behind the anchoring process in Biomercator v4.2 ?

In particular, how are the base pair distances calculated in regions where anchored markers are very far?

I have used it to retrieve genes underlying metaQTLs, but I am not really confident with the mQTL intervals when the closest anchor markers are far from the mQTL borders.

Are the intervals calculated with a ratio between centiMorgans and base pairs along the entire chromosome?

Thanks in advance

Best

Pietro

[Yannick De-Oliveira]

Dear Pietro,

 

The ratio is not really calculated along the chromosome but between two flanking markers. So, indeed the larger is the interval between two anchors, the lower will be the confidence of the projection on the physical map (PM).

Here is an example of how it calculated. Let’s say we have :

 

Anchor1 located at 44cm on GM and  at 19 000 kp on PM

Anchor2 located at 58 cm on GM and at 35 000 kp on PM

 

If you have a marker located at 48cm on your GM then the physical position is calculated this way  

 

GMdist = 58-44 = 14

PMdist = 35000-19000 = 16000

 

ProjectedPosition = ((48 – 44)/GMdist*PMdist) + 19 000 = 23 571 kbp

 

Does that makes sense ?

 

 

Best regards

Yannick

 

Yannick DE OLIVEIRA

IE Bioinformatique - INRA

Génétique Quantitative et Évolution – Le Moulon

INRA – Université Paris-Sud – CNRS – AgroParisTech

Ferme du Moulon

91190 Gif-sur-Yvette

France

+331.69.33.23.76

 

De : biome...@googlegroups.com [mailto:biome...@googlegroups.com] De la part de Pietro
Envoyé : mardi 28 mars 2017 09:34
À : biomercator <biome...@googlegroups.com>
Objet : [biomercator] how are physical intervals calculated in the anchoring process?

--
You received this message because you are subscribed to the Google Groups "biomercator" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biomercator...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Pietro]

It is clear, thanks, but how did you get to this estimates?