PACS. 82.35.Lr -Physical properties of polymers. PACS. 87.15.-v -Biomolecules: structure and physical properties.Abstract. -We present results from three-dimensional off-lattice Monte-Carlo simulations to investigate the stretching response of a semi-flexible polymer subjected to self-attractive interactions. We employ the quasi-static approximation and consider both the fixed elongation and the fixed force ensemble, which can equally well be reproduced in experiments nowadays. The force versus elongation curves are in general non-monotonic, and the quantity and height of peaks increase with decreasing temperature, and with increasing stiffness. We finally compute the variation of unfolding force with temperature. Our results should be relevant for stretching experimental studies, and for more refined theoretical modeling, taking non-equilibrium and kinetic effects into account.Introduction. -In recent years, the elastic properties of bio-polymers like proteins and DNA have been thoroughly investigated experimentally, thanks to the rapid development of single-molecule techniques, e.g. soft micro-needles [1], atomic force microscopes (AFMs) [2] and optical tweezers [3]. A major motivation for these studies comes from the possibility to enhance our quantitative understanding of the physical and biochemical properties of biomolecules. For instance, monitoring the applied tension at the ends of a macromolecule offers an insight into the molecular forces acting among its constituents. The detailed scenario emerging from the experimental force-elongation curves has called for the validating of a number of theoretical models previously introduced in polymer physics. The Freely Jointed Chain (FJC) [4] and the Worm-like Chain (WLC) models [5] are arguably the simplest and most widely known ones. These models predict markebly different shapes for the force versus elongation curves for a polymer, notably for intermediate to large forces [4,[6][7][8]. Briefly, evidence has shown that the FJC model is appropriate for flexible polymers like polyethylene, while using the WLC model is crucial when fitting elasticity data for double-stranded (ds) DNA and other semi-flexible polymers and proteins.However, neither of these models take into account of the excluded volume, and their validity is further limited to cases in which self-interactions between distinct polymer segments are negligible. This is appropriate in the strong-force regime, or far from the θ point (or folding temperature) for a self-attractive chain, such as a DNA molecule with multivalent counterions