GTC: GNN-Transformer co-contrastive learning for self-supervised heterogeneous graph representation.

Graph Neural Networks (GNNs) have emerged as the most powerful weapon for various graph tasks due to the message-passing mechanism's great local information aggregation ability. However, over-smoothing has always hindered GNNs from going deeper and capturing multi-hop neighbors. Meanwhile, most methods follow a semi-supervised learning manner, the label scarcity would limit their applicability in real-world systems.

Coaxial nanofiber scaffold with super-active platelet lysate to accelerate the repair of bone defects.

To develop biocomposite materials with the local sustained-release function of biological factors to promote bone defect repair, coaxial electrospinning technology was performed to prepare a coaxial nanofiber scaffold with super-active platelet lysate (sPL), containing gelatin/PCL/PLLA. The nanofibers exhibited a uniform bead-free round morphology, observed by a scanning electron microscope (SEM), and the core/shell structure was confirmed by a transmission electron microscope (TEM). A mixture of polycaprolactone and sPL encapsulated by hydrophilic gelatin and hydrophobic l-polylactic acid can continuously release bioactive factors for up to 40 days.