Boosting Diracs for electrophoresis.

Electrophoresis is a biochemical proces widely used in life sciences and genetics. Recovering information upon weights and proportions of atoms from electrophoresis signals may be viewed as a linear inverse question at issue in the context of a regression setup The target function to be estimated is known to be positive in the way that that a positive estimator based upon discretized Dirac functions is propos The methodology is to use boosting process for the target function with value to discretized Dirac functions. The way is illustrated on simulated data and applied to electrophoresis.

Key Words: Backfitting; Deconvolution; GCV; Merging; Positivity; Selection of Diracs; Spatial