Abstract Enrique Vidal

 

Spatial organization of the genome plays a key role in living cells. Thanks to the 3C technologies we are able to interrogate the chromatin structure. But the answers are not always easy to interpret. Data from conformation experiments present different sources of variation. Although some of them are relevant, others can be seen as undesired biases. Besides, aberrant karyotypes can distort Hi-C maps. In this situation, we aim to rule out biases before extracting useful information.

 

We present OneD, a model-based strategy that captures the main sources of unwanted variation. Used to remove biases, it improves sample reproducibility. Thanks to its properties, we can increase computational efficiency. This paves the road to a deep (high resolution) and broad (genome-wide) analysis of the genome structure. Finally, our approach can handle aneuploidies and also estimate copy number alterations.