3D printed soft surgical planning prototype for a biliary tract rhabdomyosarcoma.

Biliary tract rhabdomyosarcoma is a soft tissue malignant musculoskeletal tumor which is located in the biliary tract. Although this tumor represents less than 1% of the total amount of childhood cancers, when localized, a >70% overall 5-year survival rate, the resection is clinically challenging and complications might exist during the biliary obstruction. Although surgery remains a mainstay, complete tumor resection is generally difficult to achieve without mutilation and severe long-term sequelae. Therefore, manufacturing multi-material 3D surgical planning prototypes of the case provides a great opportunity for surgeons to learn beforehand what they can expect. Additionally, practicing before the operation enhances the probability of success. That is why different compositions of materials have been characterized to match the mechanical properties of the liver. To do this, Dynamic Mechanical Analysis (DMA) tests and Shore hardness tests have been carried out. Amongst the material samples produced, 6%wt PVA (poly vinyl alcohol)/1%wt PHY (Phytagel)-1FT (Freeze-Thaw cycles) and 1%wt agarose appear as the best options for mimicking the liver tissue in terms of viscoelasticity. Regarding the Shore hardness, the best solution is 1%wt agarose. Additionally, a surgical planning prototype using this last material mentioned was manufactured and validated using a CT (Computed Tomography) scanner. In most of the structures the difference between the 3D model and the organ in terms of dimensions is less than 3.35 mm, which represents a low dimensional error, around 1%. On the other hand, the total manufacturing cost of the 3D physical model was €513 which is relatively low in comparison with other technologies.

Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning.

Predicting interactions between proteins and other biomolecules solely based on structure remains a challenge in biology. A high-level representation of protein structure, the molecular surface, displays patterns of chemical and geometric features that fingerprint a protein's modes of interactions with other biomolecules. We hypothesize that proteins participating in similar interactions may share common fingerprints, independent of their evolutionary history. Fingerprints may be difficult to grasp by visual analysis but could be learned from large-scale datasets. We present MaSIF (molecular surface interaction fingerprinting), a conceptual framework based on a geometric deep learning method to capture fingerprints that are important for specific biomolecular interactions. We showcase MaSIF with three prediction challenges: protein pocket-ligand prediction, protein-protein interaction site prediction and ultrafast scanning of protein surfaces for prediction of protein-protein complexes. We anticipate that our conceptual framework will lead to improvements in our understanding of protein function and design.