Fig. 3: Characteristics of melt systems generated by polyply and performance.

a Atomistic structure (orange) and mappings to Martini level (blue) for all six polymer species surveyed. b End-to-end distance of the melt structures generated with polyply compared to those obtained by theory using the HRM (left) and WCM (right). Blue squares indicate Martini structures, and orange diamonds GROMOS. For each of the polymers 10 systems with four different chain lengths (50, 100, 250, 500) were built. c Single chain of atomistic Polystyrene with 100 residues shown in the melt, other chains are shown in gray and residues within 1 nm are omitted for clarity The backbone of the chain is highlighted in orange and the side chains in yellow. d Typical time for generating coordinates with polyply for different total numbers of residues. The performance is shown per force-field (blue: Martini3, orange: GROMOS), averaged over all systems in panel b leading to a total sample size of n = 60 for all systems with less than 1 million residues. In addition, systems with 1 million residues have also been set up as a further benchmark, however, considering only five replicas (i.e. n = 5). The boxplots are shown as the median within the bounding box corresponding to Interquartile-range (IQR) (Q3–Q1) and Tukey-style whiskers extending to 1.5 IQR. Open circles are outliers and in the case of atomistic melt systems, the outliers are all from melt systems of Polyethylene. Source data are provided as a source data file.