Artificial Intelligence in Health Care: Current Applications and Issues.

In recent years, artificial intelligence (AI) technologies have greatly advanced and become a reality in many areas of our daily lives. In the health care field, numerous efforts are being made to implement the AI technology for practical medical treatments. With the rapid developments in machine learning algorithms and improvements in hardware performances, the AI technology is expected to play an important role in effectively analyzing and utilizing extensive amounts of health and medical data. However, the AI technology has various unique characteristics that are different from the existing health care technologies. Subsequently, there are a number of areas that need to be supplemented within the current health care system for the AI to be utilized more effectively and frequently in health care. In addition, the number of medical practitioners and public that accept AI in the health care is still low; moreover, there are various concerns regarding the safety and reliability of AI technology implementations. Therefore, this paper aims to introduce the current research and application status of AI technology in health care and discuss the issues that need to be resolved.

Rewiring of PDZ domain-ligand interaction network contributed to eukaryotic evolution.

PDZ domain-mediated interactions have greatly expanded during metazoan evolution, becoming important for controlling signal flow via the assembly of multiple signaling components. The evolutionary history of PDZ domain-mediated interactions has never been explored at the molecular level. It is of great interest to understand how PDZ domain-ligand interactions emerged and how they become rewired during evolution. Here, we constructed the first human PDZ domain-ligand interaction network (PDZNet) together with binding motif sequences and interaction strengths of ligands. PDZNet includes 1,213 interactions between 97 human PDZ proteins and 591 ligands that connect most PDZ protein-mediated interactions (98%) in a large single network via shared ligands. We examined the rewiring of PDZ domain-ligand interactions throughout eukaryotic evolution by tracing changes in the C-terminal binding motif sequences of the PDZ ligands. We found that interaction rewiring by sequence mutation frequently occurred throughout evolution, largely contributing to the growth of PDZNet. The rewiring of PDZ domain-ligand interactions provided an effective means of functional innovations in nervous system development. Our findings provide empirical evidence for a network evolution model that highlights the rewiring of interactions as a mechanism for the development of new protein functions. PDZNet will be a valuable resource to further characterize the organization of the PDZ domain-mediated signaling proteome.

Molecular evolution of protein conformational changes revealed by a network of evolutionarily coupled residues.

An improved understanding of protein conformational changes has broad implications for elucidating the mechanisms of various biological processes and for the design of protein engineering experiments. Understanding rearrangements of residue interactions is a key component in the challenge of describing structural transitions. Evolutionary properties of protein sequences and structures are extensively studied; however, evolution of protein motions, especially with respect to interaction rearrangements, has yet to be explored. Here, we investigated the relationship between sequence evolution and protein conformational changes and discovered that structural transitions are encoded in amino acid sequences as coevolving residue pairs. Furthermore, we found that highly coevolving residues are clustered in the flexible regions of proteins and facilitate structural transitions by forming and disrupting their interactions cooperatively. Our results provide insight into the evolution of protein conformational changes and help to identify residues important for structural transitions.

ConPlex: a server for the evolutionary conservation analysis of protein complex structures.

Evolutionary conservation analyses are important for the identification of protein-protein interactions. For protein complex structures, sequence conservation has been applied to determine protein oligomerization states, to characterize native interfaces from non-specific crystal contacts, and to discriminate near-native structures from docking artifacts. However, a user-friendly web-based service for evolutionary conservation analysis of protein complexes has not been available. Therefore, we developed ConPlex (http://sbi.postech.ac.kr/ConPlex/) a web application that enables evolutionary conservation analyses of protein interactions within protein quaternary structures. Users provide protein complex structures; ConPlex automatically identifies protein interfaces and carries out evolutionary conservation analyses for the interface regions. Moreover, ConPlex allows the results of the residue-specific conservation analysis to be displayed on the protein complex structure and provides several options to customize the display output to fit each user's needs. We believe that ConPlex offers a convenient platform to analyze protein complex structures based on evolutionary conservation of protein-protein interface residues.

Comparative analysis of the secretory proteome of human adipose stromal vascular fraction cells during adipogenesis.

Adipogenesis is a complex process that is accompanied by a number of molecular events. In this study, a proteomic approach was adopted to identify secretory factors associated with adipogenesis. A label-free shotgun proteomic strategy was implemented to analyze proteins secreted by human adipose stromal vascular fraction cells and differentiated adipocytes. A total of 474 proteins were finally identified and classified according to quantitative changes and statistical significances. Briefly, 177 proteins were significantly upregulated during adipogenesis (Class I), whereas 60 proteins were significantly downregulated (Class II). Changes in the expressions of several proteins were confirmed by quantitative RT-PCR and immunoblotting. One obvious finding based on proteomic data was that the amounts of several extracellular modulators of Wnt and transforming growth factor-beta (TGF-beta) signaling changed during adipogenesis. The expressions of secreted frizzled-related proteins, dickkopf-related proteins, and latent TGF-beta-binding proteins were found to be altered during adipogenesis, which suggests that they participate in the fine regulation of Wnt and TGF-beta signaling. This study provides useful tools and important clues regarding the roles of secretory factors during adipogenic differentiation, and provides information related to obesity and obesity-related metabolic diseases.

The roles of phospholipase D in EGFR signaling.

Epidermal growth factor receptor (EGFR) is a representative model of receptor tyrosine kinases (RTKs), and offers a means of understanding their common principles and fundamental mechanisms. Furthermore, EGFR plays an essential role in cell proliferation and migration, and the disruption of EGFR signaling has been implicated in the development and growth of cancer. Phospholipase D (PLD) is a key mediator of EGFR function, and can be directly regulated by upstream binding partners in an EGF-dependent manner. PLD regulates downstream molecules by generating phosphatidic acid (PA), but it also dynamically interacts with a variety of intracellular molecules and these interactions spatiotemporally regulate EGFR function and serve as a hub that orchestrates signaling flow. This review summarizes the interrelationship between PLD and its binding molecules in the context of EGFR signaling, and addresses the roles of PLD in the mediation and coordination of this signaling.

Comparative proteomic analysis of the insulin-induced L6 myotube secretome.

Emerging evidence has revealed an endocrine function for skeletal muscle; in fact, certain anti-inflammatory cytokines are secreted only from contractile skeletal muscle. However, the skeletal muscle secretome as a whole is poorly characterized, as is how it changes in response to extracellular stimuli. Herein, we sought to identify and characterize the members of the skeletal muscle secretome, and to determine which protein secretion levels were modulated in response to insulin stimulation. To conduct these studies, we treated differentiated L6 rat skeletal muscle cells with insulin or left them untreated, and we comparatively analyzed the proteins secreted into the media. We fractionated this conditioned media using offline RP HPLC, digested the fractionated proteins, and analyzed the resulting peptides with LC-ESI-MS/MS. We identified a total of 254 proteins, and by using three different filtering methods, we identified 153 of these as secretory proteins. Fourteen proteins were secreted at higher levels under insulin stimulation, including several proteins known to be highly secreted in metabolic diseases; 19 proteins were secreted at lower levels under insulin stimulation. These result not only pinpointed several previously unknown, insulin induced, secretory proteins of skeletal muscle, it also described a novel approach for conditioned secretome analysis.

Evolutionary conservation in multiple faces of protein interaction.

Protein interfaces are believed to be evolutionarily more conserved than the rest of the protein surface, but this has not been properly verified using a large protein structural set. Furthermore, recent systematic protein interaction analyses have proved that proteins interacting with many partners have multiple interfaces to connect protein interaction networks, which have never taken into account for conservation analysis of protein interface. Here, we studied the evolutionary conservation of protein interfaces using a large-scale dataset of 2646 protein interfaces with the classification of homodimeric/heterodimeric and obligatory/transient interactions, considering all their known multiple interfaces. We found that protein interfaces were indeed more conserved than noninterface surfaces, and the conservation level of protein interfaces increased when multiple interfaces were properly considered. These findings suggest that conservation analysis should be a good descriptor for protein interface identification and protein-protein interaction predictions. We applied this evolutionary feature to filter docking decoys and found that protein interface conservation worked remarkably well in selecting the near-native structures from the large number of generated docking complexes. Moreover, we discovered that a strong correlation exist between protein interface size and protein interface conservation, which could be a useful filter for the prediction of protein-protein interactions.

Computational design of binding proteins to EGFR domain II.

We developed a process to produce novel interactions between two previously unrelated proteins. This process selects protein scaffolds and designs protein interfaces that bind to a surface patch of interest on a target protein. Scaffolds with shapes complementary to the target surface patch were screened using an exhaustive computational search of the human proteome and optimized by directed evolution using phage display. This method was applied to successfully design scaffolds that bind to epidermal growth factor receptor (EGFR) domain II, the interface of EGFR dimerization, with high reactivity toward the target surface patch of EGFR domain II. One potential application of these tailor-made protein interactions is the development of therapeutic agents against specific protein targets.

Robust, tunable biological oscillations from interlinked positive and negative feedback loops.

A simple negative feedback loop of interacting genes or proteins has the potential to generate sustained oscillations. However, many biological oscillators also have a positive feedback loop, raising the question of what advantages the extra loop imparts. Through computational studies, we show that it is generally difficult to adjust a negative feedback oscillator's frequency without compromising its amplitude, whereas with positive-plus-negative feedback, one can achieve a widely tunable frequency and near-constant amplitude. This tunability makes the latter design suitable for biological rhythms like heartbeats and cell cycles that need to provide a constant output over a range of frequencies. Positive-plus-negative oscillators also appear to be more robust and easier to evolve, rationalizing why they are found in contexts where an adjustable frequency is unimportant.