Search NASA⌕ Search

SEARCH · Search NASA

Results for “Network interaction”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 127 records · Page 7

Dynamic Control of Synaptic Adhesion and Organizing Molecules in Synaptic Plasticity

Synapses play a critical role in establishing and maintaining neural circuits, permitting targeted information transfer throughout the brain. A large portfolio of synaptic adhesion/organizing molecules (SAMs) exists in the mammalian brain involved in synapse development and maintenance. SAMs bind protein partners, forming trans -complexes spanning the synaptic cleft or cis -complexes attached to the same synaptic membrane. SAMs play key roles in cell adhesion and in organizing protein interaction networks; they can also provide mechanisms of recognition, generate scaffolds onto which partners can dock, and likely take part in signaling processes as well. SAMs are regulated through a portfolio of different mechanisms that affect their protein levels, precise localization, stability, and the availability of their partners at synapses. Interaction of SAMs with their partners can further be strengthened or weakened through alternative splicing, competing protein partners, ectodomain shedding, or astrocytically secreted factors. Given that numerous SAMs appear altered by synaptic activity, in vivo, these molecules may be used to dynamically scale up or scale down synaptic communication. Many SAMs, including neurexins, neuroligins, cadherins, and contactins, are now implicated in neuropsychiatric and neurodevelopmental diseases, such as autism spectrum disorder, schizophrenia, and bipolar disorder and studying their molecular mechanisms holds promise for developing novel therapeutics.

60 APPLIED LIFE SCIENCES↗

Separating Facts from Fiction: Linguistic Models to Classify Suspicious and Trusted News Posts on Twitter

Pew research polls report 62 percent of U.S. adults get news on social media (Gottfried and Shearer, 2016). In a December poll, 64 percent of U.S. adults said that “made-up news” has caused a “great deal of confusion” about the facts of current events (Barthel et al., 2016). Fabricated stories spread in social media, ranging from deliberate propaganda to hoaxes and satire, contributes to this confusion in addition to having serious effects on global stability. In this work we build predictive models to classify 130 thousand news tweets as suspicious or verified, and predict four subtypes of suspicious news – satire, hoaxes, clickbait and propaganda. We demonstrate that neural network models trained on tweet content and social network interactions outperform lexical models. Unlike previous work on deception detection, we find that adding syntax and grammar features to our models decreases performance. Incorporating linguistic features, including bias and subjectivity, improves classification results, however social interaction features are most informative for finer-grained separation between our four types of suspicious news posts.

classification↗

Discovery of Protein–lncRNA Interactions by Integrating Large-Scale CLIP-Seq and RNA-Seq Datasets

Long non-coding RNAs (lncRNAs) are emerging as important regulatory molecules in developmental, physiological, and pathological processes. However, the precise mechanism and functions of most of lncRNAs remain largely unknown. Recent advances in high-throughput sequencing of immunoprecipitated RNAs after cross-linking (CLIP-Seq) provide powerful ways to identify biologically relevant protein–lncRNA interactions. In this study, by analyzing millions of RNA-binding protein (RBP) binding sites from 117 CLIP-Seq datasets generated by 50 independent studies, we identified 22,735 RBP–lncRNA regulatory relationships. We found that one single lncRNA will generally be bound and regulated by one or multiple RBPs, the combination of which may coordinately regulate gene expression. We also revealed the expression correlation of these interaction networks by mining expression profiles of over 6000 normal and tumor samples from 14 cancer types. Our combined analysis of CLIP-Seq data and genome-wide association studies data discovered hundreds of disease-related single nucleotide polymorphisms resided in the RBP binding sites of lncRNAs. Finally, we developed interactive web implementations to provide visualization, analysis, and downloading of the aforementioned large-scale datasets. Our study represented an important step in identification and analysis of RBP–lncRNA interactions and showed that these interactions may play crucial roles in cancer and genetic diseases.

60 APPLIED LIFE SCIENCES↗

HDX–MS finds that partial unfolding with sequential domain activation controls condensation of a cellular stress marker

Eukaryotic cells form condensates to sense and adapt to their environment [S. F. Banani, H. O. Lee, A. A. Hyman, M. K. Rosen,Nat. Rev. Mol. Cell Biol.18, 285–298 (2017), H. Yoo, C. Triandafillou, D. A. Drummond,J. Biol. Chem.294, 7151–7159 (2019)]. Poly(A)-binding protein (Pab1), a canonical stress granule marker, condenses upon heat shock or starvation, promoting adaptation [J. A. Ribacket al.,Cell168, 1028–1040.e19 (2017)]. The molecular basis of condensation has remained elusive due to a dearth of techniques to probe structure directly in condensates. We apply hydrogen–deuterium exchange/mass spectrometry to investigate the mechanism of Pab1’s condensation. Pab1’s four RNA recognition motifs (RRMs) undergo different levels of partial unfolding upon condensation, and the changes are similar for thermal and pH stresses. Although structural heterogeneity is observed, the ability of MS to describe populations allows us to identify which regions contribute to the condensate’s interaction network. Our data yield a picture of Pab1’s stress-triggered condensation, which we term sequential activation (Fig. 1A), wherein each RRM becomes activated at a temperature where it partially unfolds and associates with other likewise activated RRMs to form the condensate. Subsequent association is dictated more by the underlying free energy surface than specific interactions, an effect we refer to as thermodynamic specificity. Our study represents an advance for elucidating the interactions that drive condensation. Furthermore, our findings demonstrate how condensation can use thermodynamic specificity to perform an acute response to multiple stresses, a potentially general mechanism for stress-responsive proteins.

Science & Technology - Other Topics↗

Discovery of Chemical Toxicity via Biological Networks and Systems Biology

Both soldiers and animals are exposed to many chemicals as the result of military activities. Tools are needed to understand the hazards and risks that chemicals and new materials pose to soldiers and the environment. We have investigated the potential of global gene regulatory networks in understanding the impact of chemicals on reproduction. We characterized effects of chemicals on ovaries of the model animal system, the Fathead minnow (Pimopheles promelas) connecting chemical impacts on gene expression to circulating blood levels of the hormones testosterone and estradiol in addition to the egg yolk protein vitellogenin. We describe the application of reverse engineering complex interaction networks from high dimensional gene expression data to characterize chemicals that disrupt the hypothalamus-pituitary-gonadal endocrine axis that governs reproduction in fathead minnows. The construction of global gene regulatory networks provides deep insights into how drugs and chemicals effect key organs and biological pathways.

59 BASIC BIOLOGICAL SCIENCES↗

Structure of the SPRY domain of the human RNA helicase DDX1, a putative interaction platform within a DEAD-box protein

The structure of the SPRY domain of the human RNA helicase DDX1 was determined at 2.0 Å resolution. The SPRY domain provides a putative protein–protein interaction platform within DDX1 that differs from other SPRY domains in its structure and conserved regions. The human RNA helicase DDX1 in the DEAD-box family plays an important role in RNA processing and has been associated with HIV-1 replication and tumour progression. Whereas previously described DEAD-box proteins have a structurally conserved core, DDX1 shows a unique structural feature: a large SPRY-domain insertion in its RecA-like consensus fold. SPRY domains are known to function as protein–protein interaction platforms. Here, the crystal structure of the SPRY domain of human DDX1 (hDSPRY) is reported at 2.0 Å resolution. The structure reveals two layers of concave, antiparallel β-sheets that stack onto each other and a third β-sheet beneath the β-sandwich. A comparison with SPRY-domain structures from other eukaryotic proteins showed that the general β-sandwich fold is conserved; however, differences were detected in the loop regions, which were identified in other SPRY domains to be essential for interaction with cognate partners. In contrast, in hDSPRY these loop regions are not strictly conserved across species. Interestingly, though, a conserved patch of positive surface charge is found that may replace the connecting loops as a protein–protein interaction surface. The data presented here comprise the first structural information on DDX1 and provide insights into the unique domain architecture of this DEAD-box protein. By providing the structure of a putative interaction domain of DDX1, this work will serve as a basis for further studies of the interaction network within the hetero-oligomeric complexes of DDX1 and of its recruitment to the HIV-1 Rev protein as a viral replication factor.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Analyses of GWAS signal using GRIN identify additional genes contributing to suicidal behavior

Genome-wide association studies (GWAS) identify genetic variants underlying complex traits but are limited by stringent genome-wide significance thresholds. We present GRIN (Gene set Refinement through Interacting Networks), which increases confidence in the expanded gene set by retaining genes strongly connected by biological networks when GWAS thresholds are relaxed. GRIN was validated on both simulated interrelated gene sets as well as multiple GWAS traits. From multiple GWAS summary statistics of suicide attempt, a complex phenotype, GRIN identified additional genes that replicated across independent cohorts and retained biologically interrelated genes despite a relaxed significance threshold. We present a conceptual model of how these retained genes interact through neurobiological pathways that may influence suicidal behavior, and identify existing drugs associated with these pathways that would not have been identified under traditional GWAS thresholds. We demonstrate GRIN’s utility in boosting GWAS results by increasing the number of true positive genes identified from GWAS results.

60 APPLIED LIFE SCIENCES↗

Power grid reliability and security

This project has led to the development of a real-time simulation platform for electric power grids called Grid Simulator or GridSim for simulating the dynamic and information network interactions of large- scale power systems. The platform consists of physical models of power system components including synchronous generators, loads and control, which are simulated using a modified commercial power simulator namely Transient Stability Analysis Tool (TSAT) [1] together with data cleanup components, as well as an emulated substation level and wide-area power analysis components. The platform also includes realistic representations of communication network middleware that can emulate the real-time information flow back and forth between substations and control centers in wide-area power systems. The platform has been validated on a realistic 6000-bus model of the western American power system. The simulator GridSim developed in this project is the first of its kind in its ability to simulate real-time response of large-scale power grids, and serves as a cost effective real-time stability and control simulation platform for power industry.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Structural basis of transcription: RNA polymerase II substrate binding and metal coordination using a free-electron laser

Catalysis and translocation of multisubunit DNA-directed RNA polymerases underlie all cellular mRNA synthesis. RNA polymerase II (Pol II) synthesizes eukaryotic pre-mRNAs from a DNA template strand buried in its active site. Structural details of catalysis at near-atomic resolution and precise arrangement of key active site components have been elusive. Here, we present the free-electron laser (FEL) structures of a matched ATP-bound Pol II and the hyperactive Rpb1 T834P bridge helix (BH) mutant at the highest resolution to date. The radiation-damage-free FEL structures reveal the full active site interaction network, including the trigger loop (TL) in the closed conformation, bonafide occupancy of both site A and B Mg 2+ , and, more importantly, a putative third (site C) Mg 2+ analogous to that described for some DNA polymerases but not observed previously for cellular RNA polymerases. Molecular dynamics (MD) simulations of the structures indicate that the third Mg 2+ is coordinated and stabilized at its observed position. TL residues provide half of the substrate binding pocket while multiple TL/BH interactions induce conformational changes that could allow translocation upon substrate hydrolysis. Consistent with TL/BH communication, a FEL structure and MD simulations of the T834P mutant reveal rearrangement of some active site interactions supporting potential plasticity in active site function and long-distance effects on both the width of the central channel and TL conformation, likely underlying its increased elongation rate at the expense of fidelity.

Science & Technology - Other Topics↗

Rapid interactome profiling by massive sequencing

We have developed a high-throughput protein expression and interaction analysis platform that combines cDNA phage display library selection and massive gene sequencing using the 454 platform. A phage display library of open reading frame (ORF) fragments was created from mRNA derived from different tissues. This was used to study the interaction network of the enzyme transglutaminase 2 (TG2), a multifunctional enzyme involved in the regulation of cell growth, differentiation and apoptosis, associated with many different pathologies. After two rounds of panning with TG2 we assayed the frequency of ORFs within the selected phage population using 454 sequencing. Ranking and analysis of more than 120 000 sequences allowed us to identify several potential interactors, which were subsequently confirmed in functional assays. Within the identified clones, three had been previously described as interacting proteins (fibronectin, SMOC1 and GSTO2), while all the others were new. When compared with standard systems, such as microtiter enzyme-linked immunosorbant assay, the method described here is dramatically faster and yields far more information about the interaction under study, allowing better characterization of complex systems. For example, in the case of fibronectin, it was possible to identify the specific domains involved in the interaction.

59 BASIC BIOLOGICAL SCIENCES↗

Differential Network Analyses of Alzheimer’s Disease Identify Early Events in Alzheimer’s Disease Pathology

In late-onset Alzheimer’s disease (AD), multiple brain regions are not affected simultaneously. Comparing the gene expression of the affected regions to identify the differences in the biological processes perturbed can lead to greater insight into AD pathogenesis and early characteristics. We identified differentially expressed (DE) genes from single cell microarray data of four AD affected brain regions: entorhinal cortex (EC), hippocampus (HIP), posterior cingulate cortex (PCC), and middle temporal gyrus (MTG). We organized the DE genes in the four brain regions into region-specific gene coexpression networks. Differential neighborhood analyses in the coexpression networks were performed to identify genes with low topological overlap (TO) of their direct neighbors. The low TO genes were used to characterize the biological differences between two regions. Our analyses show that increased oxidative stress, along with alterations in lipid metabolism in neurons, may be some of the very early events occurring in AD pathology. Cellular defense mechanisms try to intervene but fail, finally resulting in AD pathology as the disease progresses. Furthermore, disease annotation of the low TO genes in two independent protein interaction networks has resulted in association between cancer, diabetes, renal diseases, and cardiovascular diseases.

60 APPLIED LIFE SCIENCES↗

circRNA expression analysis in lung adenocarcinoma: comparison of paired fresh frozen and formalin-fixed paraffin-embedded specimens

Highlights: • First comparison of circRNA expression from paired fresh frozen and FFPE specimens. • Good correlation in circRNAs expression data utilizing fresh frozen or FFPE tissues. • Samples clustered correctly by differentially expressed circRNAs in either group. • Three differentially expressed circRNAs in lung cancers verified by qRT-PCR assay. • circRNA-miRNA-protein coding gene regulation network analyzed by bioinformatics tools. Circular RNA (circRNAs) is a novel class of endogenous non-coding RNAs which is widely involved in carcinogenesis. Archived formalin-fixed paraffin-embedded (FFPE) specimens represent valuable resources for cancer research. Currently there is a lack of systematic analysis on the feasibility of circRNAs expression analysis using FFPE samples. Here, we reported the first comparison study of circRNA expression from paired fresh frozen and FFPE specimens of lung adenocarcinoma. circRNAs expression of paired lung adenocarcinoma and adjacent normal samples, starting from either fresh frozen or FFPE materials, were analyzed via a high-throughput circRNA microarray. Hierarchical clustering was performed on samples. qRT-PCR was used to confirm the differentially expressed circRNAs. The circRNA regulation networks were built by bioinformatics tools for several candidate circRNAs. Our results demonstrated that there was a good correlation in circRNAs expression analysis utilizing fresh frozen or FFPE tissues. Tumors and adjacent normal tissues can be clustered correctly by the differentially expressed circRNAs in fresh frozen or FFPE groups. Furthermore, three differentially expressed circRNAs, hsa{sub c}ircRNA{sub 1}00421, hsa{sub c}ircRNA{sub 1}04329 and hsa{sub c}ircRNA{sub 1}01969 were verified by qRT-PCR assay. Bioinformatics analysis was also applied to predict the circRNA-miRNA-protein coding gene interaction network for each of above circRNAs. To the best of our knowledge, this is the first report demonstrating that the circRNAs microarray analysis, starting from FFPE specimens, is comparable with that based on fresh frozen samples. Therefore FFPE specimen represents a good substitute for fresh frozen tissue, especially when fresh frozen specimen is limited.

60 APPLIED LIFE SCIENCES↗

Structural and Biochemical Characterization of SrcA, a Multi-cargo Type III Secretion Chaperone in Salmonella Required for Pathogenic Association with a Host

Many Gram-negative bacteria colonize and exploit host niches using a protein apparatus called a type III secretion system (T3SS) that translocates bacterial effector proteins into host cells where their functions are essential for pathogenesis. A suite of T3SS-associated chaperone proteins bind cargo in the bacterial cytosol, establishing protein interaction networks needed for effector translocation into host cells. In Salmonella enterica serovar Typhimurium, a T3SS encoded in a large genomic island (SPI-2) is required for intracellular infection, but the chaperone complement required for effector translocation by this system is not known. Using a reverse genetics approach, we identified a multi-cargo secretion chaperone that is functionally integrated with the SPI-2-encoded T3SS and required for systemic infection in mice. Crystallographic analysis of SrcA at a resolution of 2.5 {angstrom} revealed a dimer similar to the CesT chaperone from enteropathogenic E. coli but lacking a 17-amino acid extension at the carboxyl terminus. Further biochemical and quantitative proteomics data revealed three protein interactions with SrcA, including two effector cargos (SseL and PipB2) and the type III-associated ATPase, SsaN, that increases the efficiency of effector translocation. Using competitive infections in mice we show that SrcA increases bacterial fitness during host infection, highlighting the in vivo importance of effector chaperones for the SPI-2 T3SS.

59 BASIC BIOLOGICAL SCIENCES↗

Ecosystem consequences of introducing plant growth promoting rhizobacteria to managed systems and potential legacy effects

The rapidly growing industry of crop biostimulants leverages the application of plant growth promoting rhizobacteria (PGPR) to promote plant growth and health. However, introducing nonnative rhizobacteria may impact other aspects of ecosystem functioning and have legacy effects; these potential consequences are largely unexplored. Nontarget consequences of PGPR may include changes in resident microbiomes, nutrient cycling, pollinator services, functioning of other herbivores, disease suppression, and organic matter persistence. Importantly, we lack knowledge of whether these ecosystem effects may manifest in adjacent ecosystems. The introduced PGPR can leave a functional legacy whether they persist in the community or not. Legacy effects include shifts in resident microbiomes and their temporal dynamics, horizontal transfer of genes from the PGPR to resident taxa, and changes in resident functional groups and interaction networks. Ecosystem functions may be affected by legacies PGPR leave following niche construction, such as when PGPR alter soil pH that in turn alters biogeochemical cycling rates. Here, we highlight new research directions to elucidate how introduced PGPR impact resident microbiomes and ecosystem functions and their capacity for legacy effects.

59 BASIC BIOLOGICAL SCIENCES↗

Molecular insight into conformational transmission of human P-glycoprotein

P-glycoprotein (P-gp), a kind of ATP-binding cassette transporter, can export candidates through a channel at the two transmembrane domains (TMDs) across the cell membranes using the energy released from ATP hydrolysis at the two nucleotide-binding domains (NBDs). Considerable evidence has indicated that human P-gp undergoes large-scale conformational changes to export a wide variety of anti-cancer drugs out of the cancer cells. However, molecular mechanism of the conformational transmission of human P-gp from the NBDs to the TMDs is still unclear. Herein, targeted molecular dynamics simulations were performed to explore the atomic detail of the conformational transmission of human P-gp. It is confirmed that the conformational transition from the inward- to outward-facing is initiated by the movement of the NBDs. It is found that the two NBDs move both on the two directions (x and y). The movement on the x direction leads to the closure of the NBDs, while the movement on the y direction adjusts the conformations of the NBDs to form the correct ATP binding pockets. Six key segments (KSs) protruding from the TMDs to interact with the NBDs are identified. The relative movement of the KSs along the y axis driven by the NBDs can be transmitted through α-helices to the rest of the TMDs, rendering the TMDs to open towards periplasm in the outward-facing conformation. Twenty eight key residue pairs are identified to participate in the interaction network that contributes to the conformational transmission from the NBDs to the TMDs of human P-gp. In addition, 9 key residues in each NBD are also identified. The studies have thus provided clear insight into the conformational transmission from the NBDs to the TMDs in human P-gp.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Bioengineering and Coordination of Regulatory Networks and Intracellular Complexes to Maximize Hydrogen Production by Phototrophic Microorganisms

In this study, the Principal Investigator, F.R. Tabita has teamed up with J. C. Liao from UCLA. This project's main goal is to manipulate regulatory networks in phototrophic bacteria to affect and maximize the production of large amounts of hydrogen gas under conditions where wild-type organisms are constrained by inherent regulatory mechanisms from allowing this to occur. Unrestrained production of hydrogen has been achieved and this will allow for the potential utilization of waste materials as a feed stock to support hydrogen production. By further understanding the means by which regulatory networks interact, this study will seek to maximize the ability of currently available “unrestrained” organisms to produce hydrogen. The organisms to be utilized in this study, phototrophic microorganisms, in particular nonsulfur purple (NSP) bacteria, catalyze many significant processes including the assimilation of carbon dioxide into organic carbon, nitrogen fixation, sulfur oxidation, aromatic acid degradation, and hydrogen oxidation/evolution. Moreover, due to their great metabolic versatility, such organisms highly regulate these processes in the cell and since virtually all such capabilities are dispensable, excellent experimental systems to study aspects of molecular control and biochemistry/physiology are available.

08 HYDROGEN↗

Molecular Assemblies, Genes and Genomics Integrated Efficiently (MAGGIE)

Final report on MAGGIE. We set ambitious goals to model the functions of individual organisms and their community from molecular to systems scale. These scientific goals are driving the development of sophisticated algorithms to analyze large amounts of experimental measurements made using high throughput technologies to explain and predict how the environment influences biological function at multiple scales and how the microbial systems in turn modify the environment. By experimentally evaluating predictions made using these models we will test the degree to which our quantitative multiscale understanding wilt help to rationally steer individual microbes and their communities towards specific tasks. Towards this end we have made substantial progress towards understanding evolution of gene families, transcriptional structures, detailed structures of keystone molecular assemblies (proteins and complexes), protein interactions, biological networks, microbial interactions, and community structure. Using comparative analysis we have tracked the evolutionary history of gene functions to understand how novel functions evolve. One level up, we have used proteomics data, high-resolution genome tiling microarrays, and 5' RNA sequencing to revise genome annotations, discover new genes including ncRNAs, and map dynamically changing operon structures of five model organisms: For Desulfovibrio vulgaris Hildenborough, Pyrococcus furiosis, Sulfolobus solfataricus, Methanococcus maripaludis and Haiobacterium salinarum NROL We have developed machine learning algorithms to accurately identify protein interactions at a near-zero false positive rate from noisy data generated using tagfess complex purification, TAP purification, and analysis of membrane complexes. Combining other genome-scale datasets produced by ENIGMA (in particular, microarray data) and available from literature we have been able to achieve a true positive rate as high as 65% at almost zero false positives when applied to the manually curated training set. Applying this method to the data representing around a quarter of the fraction space for water soluble proteins in D. vulgaris, we obtained 854 reliable pair wise interactions. Further, we have developed algorithms to analyze and assign significance to protein interaction data from bait pull-down experiments and integrate these data with other systems biology data through associative biclustering in a parallel computing environment. We will 'fill-in' missing information in these interaction data using a 'Transitive Closure' algorithm and subsequently use 'Between Commonality Decomposition' algorithm to discover complexes within these large graphs of protein interactions. To characterize the metabolic activities of proteins and their complexes we are developing algorithms to deconvolute pure mass spectra, estimate chemical formula for m/z values, and fit isotopic fine structure to metabolomics data. We have discovered that in comparison to isotopic pattern fitting methods restricting the chemical formula by these two dimensions actually facilitates unique solutions for chemical formula generators. To understand how microbial functions are regulated we have developed complementary algorithms for reconstructing gene regulatory networks (GRNs). Whereas the network inference algorithms cMonkey and Inferelator developed enable de novo reconstruction of predictive models for GRNs from diverse systems biology data, the RegPrecise and RegPredict framework developed uses evolutionary comparisons of genomes from closely related organisms to reconstruct conserved regulons. We have integrated the two complementary algorithms to rapidly generate comprehensive models for gene regulation of understudied organisms. Our preliminary analyses of these reconstructed GRNs have revealed novel regulatory mechanisms and cis-regulatory motifs, as well asothers that are conserved across species. Finally, we are supporting scientific efforts in ENIGMA with data management solutions and by integrating all of the algorithms, software and data into a Knowledgebase. For instance, we have developed the RegPrecise database (http://regprecise.lbl.gov) which represents manually curated sets of regulons laying the basis for automatic annotation of regulatory interactions in closely related species. We are also in the midst of scaling up MicrobesOnline to handle the growing volume of sequence and functional genomics data. Over the last year our efforts have been focused on providing support for additional genomic and functional genomic data types. Similarly, we have developed several visualization tools to help with the exploration of complex systems biology datasets. A case in point is the Gaggle Genome Browser (GGB), which was enhanced with visualizations for plotting peptide detections and protein-DNA binding alongside transcriptome structure, plus the ability to interactively filter by signal intensity or p-value.

54 ENVIRONMENTAL SCIENCES↗

A detailed study of interpretability of deep neural network based top taggers

Abstract Recent developments in the methods of explainable artificial intelligence (XAI) allow researchers to explore the inner workings of deep neural networks (DNNs), revealing crucial information about input–output relationships and realizing how data connects with machine learning models. In this paper we explore interpretability of DNN models designed to identify jets coming from top quark decay in high energy proton–proton collisions at the Large Hadron Collider. We review a subset of existing top tagger models and explore different quantitative methods to identify which features play the most important roles in identifying the top jets. We also investigate how and why feature importance varies across different XAI metrics, how correlations among features impact their explainability, and how latent space representations encode information as well as correlate with physically meaningful quantities. Our studies uncover some major pitfalls of existing XAI methods and illustrate how they can be overcome to obtain consistent and meaningful interpretation of these models. We additionally illustrate the activity of hidden layers as neural activation pattern diagrams and demonstrate how they can be used to understand how DNNs relay information across the layers and how this understanding can help to make such models significantly simpler by allowing effective model reoptimization and hyperparameter tuning. These studies not only facilitate a methodological approach to interpreting models but also unveil new insights about what these models learn. Incorporating these observations into augmented model design, we propose the particle flow interaction network model and demonstrate how interpretability-inspired model augmentation can improve top tagging performance.

97 MATHEMATICS AND COMPUTING↗