SUAG Single-Use Assembly Graph-Structured Dataset

The SUAG (Single-Use Assembly Graph) dataset is a comprehensive collection designed to advance research in the domain of single-use assembly systems and graph-based representations. This dataset encompasses a rich repository of structured information, bringing together detailed insights into pharmaceutical Single-Use Assemblies (SUAs) and their corresponding graph structures.

Key Features:

Single-Use Assembly Data: Detailed records of diverse SUAs, including bioreactors, tubing sets, and related components, providing a holistic view of assembly configurations.

Graph Representations: Graph structures capture the interconnectivity and relationships among components within each SUA, offering a unique perspective on their functional and spatial arrangements.

Variety of Applications: The dataset caters to a wide range of applications, from biopharmaceutical production to system optimization, facilitating diverse research endeavors.

Dataset Structure: 

The dataset containes three classes of SUA graphs, named: Bioreactor, Mixer, and conection.

Classes containe 236,196 Graphs,  1,088 Graphs, 300,327 Graphs respectively. 

Items named sequentially in each class and saved in their row nature in npy format.

Each graph file consists of a list of an edges list showing nodes' names and nodes' connectivity.

graph structure example, file m_0.npy in Mixer class: 

[['M' 'L1S1'], ['L1S1' 'L1P1'], ['M' 'L2S1'], ['L2S1' 'L2A2'], ['M' 'L5A1'], ['M' 'L3T1']]

The 'L' letter when followed by a number such as the case in L1, indicates the components level. components/nodes in different levels are not the same component/nodes. 

The node attribute/component is what follows for example in L1S1, the node/component is S. digit 1 indicates that the component S1 is not the same node as S2. 

We decided on this numerical way to reduce ambiguity when two nodes of the same components appear in the graph but they are different items in the pipeline. 

Component list:

   'B' -> 'bioreactorBag'   'H' -> 'hydrophobicFilter'   'F' -> 'hydrophilicFilter'   'Q' -> 'quickCoupler'   'Y' -> 'yConnector'   'X' -> 'xConnector'   'P' -> 'plug'   'A' -> 'asepticConnector'   'D' -> 'asepticDisconnector'   'S' -> 'straightFitting'   'M' -> 'mixerBag'   'R' -> 'couplerReducer'   'T' -> 'triclampConnector'   'N' -> 'tConnector'   'G' -> 'twoDimensionalBag'   'I' -> 'threeDimensionalBag'   'E' -> 'triclampConnector'   'O' -> 'sipConnector'   'J' -> 'bottle'   'C' -> 'mechanicDisconnector'   'Z' -> 'lConnector'

Researchers, practitioners, and educators can leverage the SUAG dataset to explore innovative solutions, develop algorithms for graph-based analyses, and enhance understanding of the evolving landscape of single-use assembly systems.