InCliniGene enables high-throughput and comprehensive in vivo clonal tracking toward clinical genomics data integration.

High-throughput clonal tracking in patients under hematopoietic stem cell gene therapy with integrating vector is instrumental in assessing bio-safety and efficacy. Monitoring the fate of millions of transplanted clones and their progeny across differentiation and proliferation over time leverages the identification of the vector integration sites, used as surrogates of clonal identity. Although γ-tracking retroviral insertion sites (γ-TRIS) is the state-of-the-art algorithm for clonal identification, the computational drawbacks in the tracking algorithm, based on a combinatorial all-versus-all strategy, limit its use in clinical studies with several thousands of samples per patient. We developed the first clonal tracking graph database, InCliniGene (https://github.com/calabrialab/InCliniGene), that imports the output files of γ-TRIS and generates the graph of clones (nodes) connected by arches if two nodes share common genomic features as defined by the γ-TRIS rules. Embedding both clonal data and their connections in the graph, InCliniGene can track all clones longitudinally over samples through data queries that fully explore the graph. This approach resulted in being highly accurate and scalable. We validated InCliniGene using an in vitro dataset, specifically designed to mimic clinical cases, and tested the accuracy and precision. InCliniGene allows extensive use of γ-TRIS in large gene therapy clinical applications and naturally realizes the full data integration of molecular and genomics data, clinical and treatment measurements and genomic annotations. Further extensions of InCliniGene with data federation and with application programming interface will support data mining toward precision, personalized and predictive medicine in gene therapy. Database URL:  https://github.com/calabrialab/InCliniGene.

SERS Investigation on Oligopeptides Used as Biomimetic Coatings for Medical Devices.

The surface-enhanced Raman scattering (SERS) spectra of three amphiphilic oligopeptides derived from EAK16 (AEAEAKAK)2 were examined to study systematic amino acid substitution effects on the corresponding interaction with Ag colloidal nanoparticles. Such self-assembling molecular systems, known as "molecular Lego", are of particular interest for their uses in tissue engineering and as biomimetic coatings for medical devices because they can form insoluble macroscopic membranes under physiological conditions. Spectra were collected for both native and gamma-irradiated samples. Quantum mechanical data on two of the examined oligopeptides were also obtained to clarify the assignment of the prominent significative bands observed in the spectra. In general, the peptide-nanoparticles interaction occurs through the COO- groups, with the amide bond and the aliphatic chain close to the colloid surface. After gamma irradiation, mimicking a free oxidative radical attack, the SERS spectra of the biomaterials show that COO- groups still provide the main peptide-nanoparticle interactions. However, the spatial arrangement of the peptides is different, exhibiting a systematic decrease in the distance between aliphatic chains and colloid nanoparticles.

Cloud infrastructures for in silico drug discovery: economic and practical aspects.

Cloud computing opens new perspectives for small-medium biotechnology laboratories that need to perform bioinformatics analysis in a flexible and effective way. This seems particularly true for hybrid clouds that couple the scalability offered by general-purpose public clouds with the greater control and ad hoc customizations supplied by the private ones. A hybrid cloud broker, acting as an intermediary between users and public providers, can support customers in the selection of the most suitable offers, optionally adding the provisioning of dedicated services with higher levels of quality. This paper analyses some economic and practical aspects of exploiting cloud computing in a real research scenario for the in silico drug discovery in terms of requirements, costs, and computational load based on the number of expected users. In particular, our work is aimed at supporting both the researchers and the cloud broker delivering an IaaS cloud infrastructure for biotechnology laboratories exposing different levels of nonfunctional requirements.