Important paper from Eivind Hovig's group on the new topic "Genomic DNA melting profiles"


A main research paper from Eivind Hovig's group has recently been placed online in the well reputed open access journal "Public Library of Science (PLoS) Computational Biology". It represents four years of cross-diciplinary research to open a new field of computational science, by calculating the complete melting profile of the human genome.


Brief Introduction of the Human Genomic Melting Map Project (from

The human genomic melting map is a calculated representation of cooperativity in the human genome. This implies that it provides information with respect to the probability that a DNA bubble is formed, thus providing single stranded DNA to be accessible for cellular functions. The map was established in order to provide a resource for the community to explore its biological content further.

Eivind Hovig's group has had a long standing interest in methods for mutation detection, and has published a significant body of papers on variations of denaturing gradient gel electrophoresis to this end. Presently, the methodology is developed for capillary electrophoresis using a cycling temperature system, for robust, cheap and fast assays using standard sequencing machines. This work has proven to us (and others) that prediction algorithms for DNA denaturation is well founded.

Please contact Eivind Hovig for further inquiries regarding these matters.

The melting map opens for study of for instance the correlation between flat segments and the occurence of various known biological features. Here is shown the melting map for a small area of chromosome 21, and regions the are defined as flat. Among other things, we are currently looking into the correlations of flat domains and transcription factor binding sites.
(click to enlarge image)
The figure illustrate apparent rhythms, or regular oscillations observed with the melting map along the sequence.
Two apparent rhythms are illustrated here, one with a frequency of around 150 base pairs, and one of around 500 base pairs (as indicated by the far right on the figure.
Shown below in pink are sequence repeat motifs, in black coding regions, the green represents the relative GC content, the brown curve is the actual melting map, while the small black dots along the melting map are sites of mapped single nucleotide polymorphisms. The figure illustrates a part of human chromosome 21. (click to enlarge image)



The Human Genomic Melting Map - research article published in PLoS Computational Biology, early online release (accepted April 11 2007)

Home page of Eivind Hovig's group (
Department of Tumor Biology
Institute for Cancer Research