Sequence-specific Information Research Articles

An hetero-modal deep learning framework for medical image synthesis applied to contrast and non-contrast MRI

Some pathologies such as cancer and dementia require multiple imaging modalities to fully diagnose and assess the extent of the disease. Magnetic resonance imaging offers this kind of polyvalence, but examinations take time and can require contrast agent injection. The flexible synthesis of these imaging sequences based on the available ones for a given patient could help reduce scan times or circumvent the need for contrast agent injection. In this work, we propose a deep learning architecture that can perform the synthesis of all missing imaging sequences from any subset of available images. The network is trained adversarially, with the generator consisting of parallel 3D U-Net encoders and decoders that optimally combines their multi-resolution representations with a fusion operation learned by an attention network trained conjointly with the generator network. We compare our synthesis performance with 3D networks using other types of fusion and a comparable number of trainable parameters, such as the mean/variance fusion. In all synthesis scenarios except one, the synthesis performance of the network using attention-guided fusion was better than the other fusion schemes. We also inspect the encoded representations and the attention network outputs to gain insights into the synthesis process, and uncover desirable behaviors such as prioritization of specific modalities, flexible construction of the representation when important modalities are missing, and modalities being selected in regions where they carry sequence-specific information. This work suggests that a better construction of the latent representation space in hetero-modal networks can be achieved by using an attention network.

CRISPR-Enhanced Hydrogel Microparticles for Multiplexed Detection of Nucleic Acids.

CRISPR/Cas systems offer a powerful sensing mechanism to transduce sequence-specific information into amplified analytical signals. However, performing multiplexed CRISPR/Cas assays remains challenging and often requires complex approaches for multiplexed assays. Here, a hydrogel-based CRISPR/Cas12 system termed CLAMP (Cas-Loaded Annotated Micro-Particles) is described. The approach compartmentalizes the CRISPR/Cas reaction in spatially-encoded hydrogel microparticles (HMPs). Each HMP is identifiable by its face code and becomes fluorescent when target DNA is present. The assay is further streamlined by capturing HMPs inside a microfluidic device; the captured particles are then automatically recognized by a machine-learning algorithm. The CLAMP assay is fast, highly sensitive (attomolar detection limits with preamplification), and capable of multiplexing in a single-pot assay. As a proof-of-concept clinical application, CLAMP is applied to detect nucleic acid targets of human papillomavirus in cervical brushing samples.

Adaptive Siamese Tracking with a Compact Latent Network.

In this paper, we provide an intuitive viewing to simplify the Siamese-based trackers by converting the tracking task to a classification. Under this viewing, we perform an in-depth analysis for them through visual simulations and real tracking examples, and find that the failure cases in some challenging situations can be regarded as the issue of missing decisive samples in offline training. Since the samples in the initial (first) frame contain rich sequence-specific information, we can regard them as the decisive samples to represent the whole sequence. To quickly adapt the base model to new scenes, a compact latent network is presented via fully using these decisive samples. Specifically, we present a statistics-based compact latent feature for fast adjustment by efficiently extracting the sequence-specific information. Furthermore, a new diverse sample mining strategy is designed for training to further improve the discrimination ability of the proposed compact latent network. Finally, a conditional updating strategy is proposed to efficiently update the basic models to handle scene variation during the tracking phase. To evaluate the generalization ability and effectiveness and of our method, we apply it to adjust three classical Siamese-based trackers, namely SiamRPN++, SiamFC, and SiamBAN. Extensive experimental results on six recent datasets demonstrate that all three adjusted trackers obtain the superior performance in terms of the accuracy, while having high running speed.

Structural insights into how vacuolar sorting receptors recognize the sorting determinants of seed storage proteins

In Arabidopsis, vacuolar sorting receptor isoform 1 (VSR1) sorts 12S globulins to the protein storage vacuoles during seed development. Vacuolar sorting is mediated by specific protein-protein interactions between VSR1 and the vacuolar sorting determinant located at the C terminus (ctVSD) on the cargo proteins. Here, we determined the crystal structure of the protease-associated domain of VSR1 (VSR1-PA) in complex with the C-terminal pentapeptide (468RVAAA472) of cruciferin 1, an isoform of 12S globulins. The 468RVA470 motif forms a parallel β-sheet with the switch III residues (127TMD129) of VSR1-PA, and the 471AA472 motif docks to a cradle formed by the cargo-binding loop (95RGDCYF100), making a hydrophobic interaction with Tyr99. The C-terminal carboxyl group of the ctVSD is recognized by forming salt bridges with Arg95. The C-terminal sequences of cruciferin 1 and vicilin-like storage protein 22 were sufficient to redirect the secretory red fluorescent protein (spRFP) to the vacuoles in Arabidopsis protoplasts. Adding a proline residue to the C terminus of the ctVSD and R95M substitution of VSR1 disrupted receptor-cargo interactions invitro and led to increased secretion of spRFP in Arabidopsis protoplasts. How VSR1-PA recognizes ctVSDs of other storage proteins was modeled. The last three residues of ctVSD prefer hydrophobic residues because they form a hydrophobic cluster with Tyr99 of VSR1-PA. Due to charge-charge interactions, conserved acidic residues, Asp129 and Glu132, around the cargo-binding site should prefer basic residues over acidic ones in the ctVSD. The structural insights gained may be useful in targeting recombinant proteins to the protein storage vacuoles in seeds.

3D or Not 3D: Shaping the Genome during Development.

One of the most fundamental questions in developmental biology is how one fertilized cell can give rise to a fully mature organism and how gene regulation governs this process. Precise spatiotemporal gene expression is required for development and is believed to be achieved through a complex interplay of sequence-specific information, epigenetic modifications, trans-acting factors, and chromatin folding. Here we review the role of chromatin folding during development, the mechanisms governing 3D genome organization, and how it is established in the embryo. Furthermore, we discuss recent advances and debated questions regarding the contribution of the 3D genome to gene regulation during organogenesis. Finally, we describe the mechanisms that can reshape the 3D genome, including disease-causing structural variations and the emerging view that transposable elements contribute to chromatin organization.

Validation of a Next-Generation Sequencing–Based T-Cell Receptor Gamma Gene Rearrangement Diagnostic Assay: Transitioning from Capillary Electrophoresis to Next-Generation Sequencing

Assessment of T-cell receptor γ gene (TRG) rearrangements is an importants consideration in the diagnostic workup of lymphoproliferative diseases. Although fragment analysis by PCR and capillary electrophoresis (CE) is the current standard of such assessment in clinical molecular diagnostic laboratories, it does not provide sequence information and is only semi-quantitative. Next-generation sequencing (NGS)-based assays are an attractive alternative to the conventional fragment size-based methods, given that they generate results with specific clonotype sequence information and allow for more accurate quantitation. The present study evaluated various test parameters and performance characteristics of a commercially available NGS-based TRG gene-rearrangement assay by testing 101 clinical samples previously characterized by fragment analysis. The NGS TRG assay showed an overall accuracy of 83% and an analytical specificity of 100%. The concordance rates were 88% to 95% in the Vγ1-8, Vγ10, and Vγ11 gene families, but lower in the Vγ9 gene family. This difference was mostly attributed to the incomplete polyclonal symmetry resulting from the two-tube CE assay versus the one-tube design of the NGS assay. The NGS assay also demonstrated strengths in distinguishing clonotypes of the same fragment size. This clinical validation demonstrated robust performance of the NGS-based TRG assay and identified potential pitfalls associated with CE assay design that are important for understanding the observed discrepancies with the CE-based assay.

Characterization of UVA-Induced Alterations to Transfer RNA Sequences.

Ultraviolet radiation (UVR) adversely affects the integrity of DNA, RNA, and their nucleoside modifications. By employing liquid chromatography–tandem mass spectrometry (LC–MS/MS)-based RNA modification mapping approaches, we identified the transfer RNA (tRNA) regions most vulnerable to photooxidation. Photooxidative damage to the anticodon and variable loop regions was consistently observed in both modified and unmodified sequences of tRNA upon UVA (λ 370 nm) exposure. The extent of oxidative damage measured in terms of oxidized guanosine, however, was higher in unmodified RNA compared to its modified version, suggesting an auxiliary role for nucleoside modifications. The type of oxidation product formed in the anticodon stem–loop region varied with the modification type, status, and whether the tRNA was inside or outside the cell during exposure. Oligonucleotide-based characterization of tRNA following UVA exposure also revealed the presence of novel photoproducts and stable intermediates not observed by nucleoside analysis alone. This approach provides sequence-specific information revealing potential hotspots for UVA-induced damage in tRNAs.

Outcome of Treatment After Locoregional Radiation, Neoadjuvant and Adjuvant Chemotherapy for Breast Cancer

The addition of systemic therapy to locoregional treatment for women with all stages of breast cancer improves survival. Reliable data are lacking about the outcomes of women with breast cancer in low-middle income countries who are treated with a combination of radiotherapy (RT), surgery, and systemic therapy. report the outcome of women with breast cancer treated with either neoadjuvant chemotherapy (NACT) or adjuvant chemotherapy at the National Centre for Radiotherapy and Nuclear Medicine in Accra, Ghana. The medical charts of all women with breast cancer managed at the National Centre for Radiotherapy and Nuclear Medicine from 2005 to 2014 were reviewed. Available patient specific demographic data, tumor stage, treatment sequence, and follow-up information were recorded. Logistic regression was used to estimate the presence of locoregional (LRR) and distant recurrence (DR) between two chemotherapy groups. Overall survival (OS) and disease-free survival (DFS) were used as time dependent outcomes and were analyzed using cox proportional hazard (CPH) model. All the models were adjusted with clinico-pathologic variables and estimates were reported along with their 95% confidence interval (CI). A total of 387 patients with a median follow up of 48 months were included in the study. The median age at diagnosis was 51 years. 49.6% of women had advanced pathologic stage ≥pT3. 58.9% (n = 228) received adjuvant chemotherapy whereas 41.1 % (n = 159) received NACT. The majority of women (94.1%, n = 365) received RT. 18.4% (n = 67) underwent breast conservation therapy followed by RT to a mean dose of 48.8 Gy in 10-30 fractions, and 81.6% (n = 298) had a mastectomy followed by RT to a mean dose of 47.92 Gy in 10-30 fractions. In an adjusted logistic model, no difference was observed in LRR between patients receiving NACT compared to those receiving adjuvant chemotherapy (OR = 1.05, 95% CI {0.44 - 2.47}, p = 0.9). However, NACT recipients had a higher likelihood of DR (OR = 1.97, 95% CI, {1.24 – 3.15}, p = 0.004). In a multivariate analysis, no differences were observed in survival between the two chemotherapy groups (HR = 1.43, 95% CI, {0.91 – 2.26}, p = 0.1). There was a trend towards lower DFS within the NACT group compared to the adjuvant chemotherapy group, with patients receiving NACT having a 52% likelihood of lower DFS (HR = 1.52, 95% CI, {0.96 – 2.39}, p = 0.06). While many women present with late stage disease in Ghana, outcomes after trimodal therapy are comparable to what is observed in developed countries. Advanced stage of disease is significantly associated with recurrence and survival, hence early detection in a resource limited setting is crucial to optimal outcomes. To our knowledge, this is the first and largest study to report on the outcomes of women with breast cancer who receive standard of care treatment in Ghana.

DNA-π Amphiphiles: A Unique Building Block for the Crafting of DNA-Decorated Unilamellar Nanostructures.

The unparalleled ability of DNA to recognize its complementary strand through Watson and Crick base pairing is one of the most reliable molecular recognition events found in natural systems. This highly specific sequence information encoded in DNA enables it to be a versatile building block for bottom-up self-assembly. Hence, the decoration of functional nanostructures with information-rich DNA is extremely important as this allows the integration of other functional molecules onto the surface of the nanostructures through DNA hybridization in a highly predictable manner. DNA amphiphiles are a class of molecular hybrids where a short hydrophilic DNA is conjugated to a hydrophobic moiety. Since DNA amphiphiles comprise DNA as the hydrophilic segment, their self-assembly in aqueous medium always results in the formation of nanostructures with shell made of DNA. This clearly suggests that self-assembly of DNA amphiphiles is a straightforward strategy for the ultradense decoration of a nanostructure with DNA. However, initial attempts toward the design of DNA amphiphiles were primarily focused on long flexible hydrocarbon chains as the hydrophobic moiety, and it has been demonstrated in several examples that they typically self-assemble into DNA-decorated micelles (spherical or cylindrical). Hence, molecular level control over the self-assembly of DNA amphiphiles and achieving diverse morphologies was extremely challenging and unrealized until recently.In this Account, we summarize our recent efforts in the area of self-assembly of DNA amphiphiles and narrate the remarkable effect of the incorporation of a large π-surface as the hydrophobic domain in the self-assembly of DNA amphiphiles. Self-assembly of DNA amphiphiles with flexible hydrocarbon chains as the hydrophobic moiety is primarily driven by the hydrophobic effect. The morphology of such nanostructures is typically predicted based on the volume ratio of hydrophobic to hydrophilic segments. However, control over the self-assembly and prediction of the morphology become increasingly challenging when the hydrophobic moieties can interact with each other through other noncovalent interactions. In this Account, the unique self-assembly behaviors of DNA-π amphiphiles, where a large π-surface acts as the hydrophobe, are described. Due to the extremely strong π-π stacking in aqueous medium, the assembly of the amphiphile is found to preferably proceed in a lamellar fashion (bilayer) and hence the morphology of the nanostructures can easily be tuned by the structural modification of the π-surface. Design principles for crafting various DNA-decorated lamellar nanostructures including unilamellar vesicles, two-dimensional (2D) nanosheets, and helically twisted nanoribbons by selecting suitable π-surfaces are discussed. Unilamellar vesicular nanostructures were achieved by using linear oligo(phenylene ethynylene) (OPE) as the hydrophobic segment, where lamellar assembly undergoes folding to form unilamellar vesicles. The replacement of OPE with a strongly π-stacking hydrophobe such as hexabenzocoronene (HBC) or tetraphenylethylene (TPE) provides extremely strong π-stacking compared to OPE, which efficiently directed the 2D growth for the lamellar assembly and led to the formation of 2D nanosheets. A helical twist in the lamella was achieved by the replacement of HBC with hexaphenylbenzene (HPB), which is the twisted analogue of HBC, directing the assembly into helically twisted nanoribbons. The most beneficial structural feature of this kind of nanostructure is the extremely dense decoration of their surface with ssDNA, which can further be used for DNA-directed organization of other functional nanomaterials. By exploring this, their potential as a nanoscaffold for predefined assembly of plasmonic nanomaterials into various plasmonic 1D, 2D, and 3D nanostructures through DNA hybridization is discussed. Moreover, the design of pH-responsive DNA-based vesicles and their application as a nanocarrier for payload delivery is also demonstrated.

Discrete finger sequences are widely represented in human striatum

Research in primates and rodents ascribes the striatum a critical role in integrating elementary movements into unitary action sequences through reinforcement-based learning. Yet it remains to be shown whether the human striatum represents action sequence-specific information. Young right-handed volunteers underwent functional magnetic resonance imaging while they performed four discrete finger sequences with their right hand, consisting of five button presses. Specific finger sequences could be discriminated based on the distributed activity patterns in left and right striatum, but not by average differences in single-voxel activity. Multiple bilateral clusters in putamen and caudate nucleus belonging to motor, associative, parietal and limbic territories contributed to classification sensitivity. The results show that individual finger movement sequences are widely represented in human striatum, supporting functional integration rather than segregation. The findings are compatible with the idea that the basal ganglia simultaneously integrate motor, associative and limbic aspects in the control of complex overlearned behaviour.

Visual Tracking Jointly With Online and Offline Learning

While recent approaches based on offline learning perform well in balancing the accuracy and speed of tracking, it is still non-trivial to accommodate a pre-trained model to an unseen target. In fact, online learning, which requires to capture specified target characteristics from one shot, is a unique attribute of single object tracking. How to favorably bridge the gap between the offline and online learning is of importance for tracking any unseen targets. In this work, in order to obtain sequence-specific information, we propose an online lightweight network consisting of feature adapting layer and ridge regression layer. Its key innovation is to interpret the ridge regression as one layer of the network. Furthermore, we integrate cross-similarity into the Siamese network and train it offline in an end-to-end manner to acquire the fine-grained local pattern of the target object. Through our effective fusion scheme for the offline and online procedures, our method can achieve considerable improvements on prevalent benchmarks.

A Chemometric Approach Toward Predicting the Relative Aggregation Propensity: Aβ(1-42).

A number of algorithms have been developed to predict the aggregation propensity of peptides and proteins, but virtually none have the ability to provide sequence-specific information on what physicochemical properties are most important in altering aggregation propensity. In this study, a chemometric approach using reduced amino acid properties is used to examine the aggregation behavior of a highly amyloidogenic peptide, Aβ(1-42). Specific residues are identified as being critical to the aggregation process. At each of these positions, the important physicochemical properties are identified that would either accelerate or inhibit fibril formation.

Cortical Basal Ganglia Network Interactions During Sequence Learning in Parkinson's Disease

Abstract INTRODUCTION Learning a motor skill involves organizing a series of complex movements into sequences that can be executed efficiently and reproducibly. Once learned, these sequences generate lasting changes in motor control circuits. Animal studies suggest that the interaction between the motor cortex and basal ganglia is critically involved in motor sequence learning. In particular, the cortical neurons can encode sequence-specific information that is stored subcortically once the sequence is learned. However, how motor sequence learning in humans is not well understood. In disease states like Parkinson disease, where dopaminergic denervation to the striatum affects motor functions and motor learning, understanding the circuit mechanisms of motor learning dysfunction is critical for improving motor rehabilitation. METHODS We study the neural basis of motor sequence learning in 4 Parkinson patients by performing chronic recordings of field potentials from the motor cortex (1 patient) or prefrontal cortex (3 patients) and the pallidum while patients performed the serial reaction time task (SRTT). RESULTS All patients exhibited improvements in motor sequence learning in the SRTT. There is task-modulated increase in theta (4-8 Hz) oscillations during sequence-specific trials in the motor cortex. The pallidum in all patients showed similar increases in theta oscillation at the start of motor sequences. CONCLUSION This is the first illustration of cortical basal ganglia network interactions recorded from the human brain during motor sequence learning. Increases in cortical and subcortical theta oscillations may provide a mechanism for encoding of movement sequences.

End-to-end deep metric network for visual tracking

In this paper, we propose an end-to-end deep metric network (DMN) for visual tracking, where any target can be accurately tracked given only a bounding box of the first frame. Our main motivation is to make the network learn to learn a deep distance metric by following the philosophy of one-shot learning. Instead of utilizing a hand-crafted distance metric like Euclidean distance, our DMN focuses on providing a learnable metric, which is more robust to appearance variations. Furthermore, we are the first to properly combine mean square errors and contrastive loss into a joint loss function for back-propagation. During online tracking, DMN firstly applies our instance initialization for obtaining sequence-specific information and then straightforwardly tracks the target without the help of box refinement, occlusion detection and online updating. The final tracking score considers both our DMN scalar output and the constrain of motion smoothness. Ablation analyses are carried out to validate the effectiveness of our proposed method. And experiments on the prevalent benchmarks show that our method can achieve a competitive performance when compared with some representative trackers, especially those existing metric learning-based algorithms.

CombLabel: rational design of optimized sequence-specific combinatorial labeling schemes. Application to backbone assignment of membrane proteins with low stability.

Assignment of backbone resonances is a necessary initial step in every protein NMR investigation. Standard assignment procedure is based on the set of 3D triple-resonance (1H-13C-15N) spectra and requires at least several days of experimental measurements. This limits its application to the proteins with low stability. To speed up the assignment procedure, combinatorial selective labeling (CSL) can be used. In this case, sequence-specific information is extracted from 2D spectra measured for several selectively 13C,15N-labeled samples, produced in accordance with a special CSL scheme. Here we review previous applications of the CSL approach and present novel deterministic 'CombLabel' algorithm, which generates CSL schemes minimizing the number of labeled samples and their price and maximizing assignment information that can be obtained for a given protein sequence. Theoretical calculations revealed that CombLabel software outperformed previously proposed stochastic algorithms. Current implementation of CombLabel robustly calculates CSL schemes containing up to six samples, which is sufficient for moderately sized (up to 200 residues) proteins. As a proof of concept, we calculated CSL scheme for the first voltage-sensing domain of human Nav1.4 channel, a 134 residue four helical transmembrane protein having extremely low stability in micellar solution (half-life ~ 24h at 45°C). Application of CSL doubled the extent of backbone resonance assignment, initially obtained by conventional approach. The obtained assignment coverage (~ 50%) is sufficient for ligand screening and mapping of binding interfaces.

Abstract 4444: Deep learning of the immune synapse

Abstract Artificial intelligence is poised to revolutionize every aspect of human life, finding applications in everything from self-driving cars to diagnosing cancer. In fact, almost any task that involves pattern recognition can be formulated in a way that modern AI algorithms can be used to achieve super-human performance. The immune synapse is a highly complex interaction between several proteins and peptides that allows for a constant surveillance of foreign invaders. However, modeling these interactions is extremely difficult as the combinations of interactions is simply intractable. In immuno-oncology, the study of this interaction is crucial as anti-tumor responses rely on sensitive and specific recognition of tumor-specific antigens. Implications of accurately predicting and modeling these interactions in immune-oncology range from improved and potent vaccine design to biomarkers for predicting response to immunotherapy. Our group has developed a variety of deep learning models to model the signal transmission within the immune synapse. At the core of all architectures designed, convolutional layers, similar to ones used to learn features in images, are used to learn motifs within the sequencing data for a predictive or descriptive purpose.We first present AI-MHC, an applied deep convolutional neural network for class-specific MHC binding algorithm that achieves state-of-the-art performance in both Class and Class II predictions. By incorporating ‘meaning’ of the allele within the network, we are able to model the interaction of allele and peptide within the context of a neural network. We take these concepts further in the development of DeepMANA, a deep learning framework which combines sequence-specific information about an allele/peptide pairing to not only predict binding affinity for any allele with a known protein sequence but also provide an antigen ‘quality’ score, based on the “non-self/foreign-ness” of a neoantigen. We observe in three previously published immunotherapy clinical trials, these quality neoantigens are enriched in long-term survivors or responders. Finally, we present DeepTCR, a collection of unsupervised and supervised deep learning algorithms capable of revealing structure in T-cell receptor repertoire. We demonstrate that DeepTCR achieves state-of-the-art performance in clustering antigen-specific TCR’s and is capable of learning a predictive signature in TIL repertoire of mice treated with various immunotherapies.These types of AI technologies could yield an entire new area of biomarker discovery as well as improve our understanding of the complex interaction occurring at the immune synapse that is ultimately required for a successful anti-tumor response. Citation Format: John-William Sidhom, Drew M. Pardoll, Alexander S. Baras. Deep learning of the immune synapse [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4444.

Consolidation alters motor sequence-specific distributed representations.

Functional magnetic resonance imaging (fMRI) studies investigating the acquisition of sequential motor skills in humans have revealed learning-related functional reorganizations of the cortico-striatal and cortico-cerebellar motor systems accompanied with an initial hippocampal contribution. Yet, the functional significance of these activity-level changes remains ambiguous as they convey the evolution of both sequence-specific knowledge and unspecific task ability. Moreover, these changes do not specifically assess the occurrence of learning-related plasticity. To address these issues, we investigated local circuits tuning to sequence-specific information using multivariate distances between patterns evoked by consolidated or newly acquired motor sequences production. The results reveal that representations in dorsolateral striatum, prefrontal and secondary motor cortices are greater when executing consolidated sequences than untrained ones. By contrast, sequence representations in the hippocampus and dorsomedial striatum becomes less engaged. Our findings show, for the first time in humans, that complementary sequence-specific motor representations evolve distinctively during critical phases of skill acquisition and consolidation.

Visual object tracking based on adaptive Siamese and motion estimation network

Recently, convolutional neural network (CNN) has attracted much attention in different areas of computer vision, due to its powerful abstract feature representation. Visual object tracking is one of the interesting and important areas in computer vision that achieves remarkable improvements in recent years. In this work, we aim to improve both the motion and observation models in visual object tracking by leveraging representation power of CNNs. To this end, a motion estimation network (named MEN) is utilized to seek the most likely locations of the target and prepare a further clue in addition to the previous target position. Hence the motion estimation would be enhanced by generating a small number of candidates near two plausible positions. The generated candidates are then fed into a trained Siamese network to detect the most probable candidate. Each candidate is compared to an adaptable buffer, which is updated under a predefined condition. To take into account the target appearance changes, a weighting CNN (called WCNN) adaptively assigns weights to the final similarity scores of the Siamese network using sequence-specific information. Evaluation results on well-known benchmark datasets (OTB100, OTB50 and OTB2013) prove that the proposed tracker outperforms the state-of-the-art competitors.

The Coiled-Coil NLR Rph1, Confers Leaf Rust Resistance in Barley Cultivar Sudan.

Unraveling and exploiting mechanisms of disease resistance in cereal crops is currently limited by their large repeat-rich genomes and the lack of genetic recombination or cultivar (cv)-specific sequence information. We cloned the first leaf rust resistance gene Rph1 (Rph1 a) from cultivated barley (Hordeum vulgare) using "MutChromSeq," a recently developed molecular genomics tool for the rapid cloning of genes in plants. Marker-trait association in the CI 9214/Stirling doubled haploid population mapped Rph1 to the short arm of chromosome 2H in a physical region of 1.3 megabases relative to the barley cv Morex reference assembly. A sodium azide mutant population in cv Sudan was generated and 10 mutants were confirmed by progeny-testing. Flow-sorted 2H chromosomes from Sudan (wild type) and six of the mutants were sequenced and compared to identify candidate genes for the Rph1 locus. MutChromSeq identified a single gene candidate encoding a coiled-coil nucleotide binding site Leucine-rich repeat (NLR) receptor protein that was altered in three different mutants. Further Sanger sequencing confirmed all three mutations and identified an additional two independent mutations within the same candidate gene. Phylogenetic analysis determined that Rph1 clustered separately from all previously cloned NLRs from the Triticeae and displayed highest sequence similarity (89%) with a homolog of the Arabidopsis (Arabidopsis thaliana) disease resistance protein 1 protein in Triticum urartu In this study we determined the molecular basis for Rph1-mediated resistance in cultivated barley enabling varietal improvement through diagnostic marker design, gene editing, and gene stacking technologies.

Observing Action Sequences Elicits Sequence-Specific Neural Representations in Frontoparietal Brain Regions.

Learning new skills by watching others is important for social and motor development throughout the lifespan. Prior research has suggested that observational learning shares common substrates with physical practice at both cognitive and brain levels. In addition, neuroimaging studies have used multivariate analysis techniques to understand neural representations in a variety of domains, including vision, audition, memory, and action, but few studies have investigated neural plasticity in representational space. Therefore, although movement sequences can be learned by observing other people's actions, a largely unanswered question in neuroscience is how experience shapes the representational space of neural systems. Here, across a sample of male and female participants, we combined pretraining and posttraining fMRI sessions with 6 d of observational practice to determine whether the observation of action sequences elicits sequence-specific representations in human frontoparietal brain regions and the extent to which these representations become more distinct with observational practice. Our results showed that observed action sequences are modeled by distinct patterns of activity in frontoparietal cortex and that such representations largely generalize to very similar, but untrained, sequences. These findings advance our understanding of what is modeled during observational learning (sequence-specific information), as well as how it is modeled (reorganization of frontoparietal cortex is similar to that previously shown following physical practice). Therefore, on a more fine-grained neural level than demonstrated previously, our findings reveal how the representational structure of frontoparietal cortex maps visual information onto motor circuits in order to enhance motor performance.SIGNIFICANCE STATEMENT Learning by watching others is a cornerstone in the development of expertise and skilled behavior. However, it remains unclear how visual signals are mapped onto motor circuits for such learning to occur. Here, we show that observed action sequences are modeled by distinct patterns of activity in frontoparietal cortex and that such representations largely generalize to very similar, but untrained, sequences. These findings advance our understanding of what is modeled during observational learning (sequence-specific information), as well as how it is modeled (reorganization of frontoparietal cortex is similar to that previously shown following physical practice). More generally, these findings demonstrate how motor circuit involvement in the perception of action sequences shows high fidelity to prior work, which focused on physical performance of action sequences.