Biopsy or Follow-up: AI Improves the Clinical Strategy of US BI-RADS 4A Breast Nodules Using a Convolutional Neural Network.

OBJECTIVES: To develop predictive nomograms based on clinical and ultrasound features and to improve the clinical strategy for US BI-RADS 4A lesions. METHODS: Patients with US BI-RADS 4A lesions from 3 hospitals between January 2016 and June 2020 were retrospectively included. Clinical and ultrasound features were extracted to establish nomograms CE (based on clinical experience) and DL (based on deep-learning algorithm). The performances of nomograms were evaluated by receiver operator characteristic curves, calibration curves and decision curves. Diagnostic performances with DL of radiologists were analyzed. RESULTS: 1616 patients from 2 hospitals were randomly divided into training and internal validation cohorts at a ratio of 7:3. Hundred patients from another hospital made up external validation cohort. DL achieved more optimized AUCs than CE (internal validation: 0.916 vs. 0.863, P < .01; external validation: 0.884 vs. 0.776, P = .05). The sensitivities of DL were higher than CE (internal validation: 81.03% vs. 72.41%, P = .044; external validation: 93.75% vs. 81.25%, P = .4795) without losing specificity (internal validation: 84.91% vs. 86.47%, P = .353; external validation: 69.14% vs. 71.60%, P = .789). Decision curves indicated DL adds more clinical net benefit. With DL's assistance, both radiologists achieved higher AUCs (0.712 vs. 0.801; 0.547 vs. 0.800), improved specificities (70.93% vs. 74.42%, P < .001; 59.3% vs. 81.4%, P = .004), and decreased unnecessary biopsy rates by 6.7% and 24%. CONCLUSION: DL was developed to discriminate US BI-RADS 4A lesions with a higher diagnostic power and more clinical net benefit than CE. Using DL may guide clinicians to make precise clinical decisions and avoid overtreatment of benign lesions.

Brain-wide neural activity underlying memory-guided movement.

Behavior relies on activity in structured neural circuits that are distributed across the brain, but most experiments probe neurons in a single area at a time. Using multiple Neuropixels probes, we recorded from multi-regional loops connected to the anterior lateral motor cortex (ALM), a circuit node mediating memory-guided directional licking. Neurons encoding sensory stimuli, choices, and actions were distributed across the brain. However, choice coding was concentrated in the ALM and subcortical areas receiving input from the ALM in an ALM-dependent manner. Diverse orofacial movements were encoded in the hindbrain; midbrain; and, to a lesser extent, forebrain. Choice signals were first detected in the ALM and the midbrain, followed by the thalamus and other brain areas. At movement initiation, choice-selective activity collapsed across the brain, followed by new activity patterns driving specific actions. Our experiments provide the foundation for neural circuit models of decision-making and movement initiation.

A rumen-derived bifunctional glucanase/mannanase uncanonically releases oligosaccharides with a high degree of polymerization preferentially from branched substrates.

Glycoside hydrolases (GHs) are known to depolymerize polysaccharides into oligo-/mono-saccharides, they are extensively used as additives for both animals feed and our food. Here we reported the characterization of IDSGH5-14(CD), a weakly-acidic mesophilic bifunctional mannanase/glucanase of GH5, originally isolated from sheep rumen microbes. Biochemical characterization studies revealed that IDSGH5-14(CD) exhibited preferential hydrolysis of mannan-like and glucan-like substrates. Interestingly, the enzyme exhibited significantly robust catalytic activity towards branched-substrates compared to linear polysaccharides (P < 0.05). Substrate hydrolysis pattern indicated that IDSGH5-14(CD) predominantly liberated oligosaccharides with a degree of polymerization (DP) of 3-7 as the end products, dramatically distinct from canonical endo-acting enzymes. Comparative modeling revealed that IDSGH5-14(CD) was mainly comprised of a (ß/α)8-barrel-like structure with a spacious catalytic cleft on surface, facilitating the enzyme to target high-DP or branched oligosaccharides. Molecular dynamics (MD) simulations further suggested that the branched-ligand, 64-α-D-galactosyl-mannohexose, was steadily accommodated within the catalytic pocket via a two-sided clamp formed by the aromatic residues. This study first reports a bifunctional GH5 enzyme that predominantly generates high-DP oligosaccharides, preferentially from branched-substrates. This provides novel insights into the catalytic mechanism and molecular underpinnings of polysaccharide depolymerization, with potential implications for feed additive development and high-DP oligosaccharides preparation.

Superior colliculus bidirectionally modulates choice activity in frontal cortex.

Action selection occurs through competition between potential choice options. Neural correlates of choice competition are observed across frontal cortex and downstream superior colliculus (SC) during decision-making, yet how these regions interact to mediate choice competition remains unresolved. Here we report that SC can bidirectionally modulate choice competition and drive choice activity in frontal cortex. In the mouse, topographically matched regions of frontal cortex and SC formed a descending motor pathway for directional licking and a re-entrant loop via the thalamus. During decision-making, distinct neuronal populations in both frontal cortex and SC encoded opposing lick directions and exhibited competitive interactions. SC GABAergic neurons encoded ipsilateral choice and locally inhibited glutamatergic neurons that encoded contralateral choice. Activating or suppressing these cell types could bidirectionally drive choice activity in frontal cortex. These results thus identify SC as a major locus to modulate choice competition within the broader action selection network.

Activity map of a cortico-cerebellar loop underlying motor planning.

The neocortex and cerebellum interact to mediate cognitive functions. It remains unknown how the two structures organize into functional networks to mediate specific behaviors. Here we delineate activity supporting motor planning in relation to the mesoscale cortico-cerebellar connectome. In mice planning directional licking based on short-term memory, preparatory activity instructing future movement depends on the anterior lateral motor cortex (ALM) and the cerebellum. Transneuronal tracing revealed divergent and largely open-loop connectivity between the ALM and distributed regions of the cerebellum. A cerebellum-wide survey of neuronal activity revealed enriched preparatory activity in hotspot regions with conjunctive input-output connectivity to the ALM. Perturbation experiments show that the conjunction regions were required for maintaining preparatory activity and correct subsequent movement. Other cerebellar regions contributed little to motor planning despite input or output connectivity to the ALM. These results identify a functional cortico-cerebellar loop and suggest the cerebellar cortex selectively establishes reciprocal cortico-cerebellar communications to orchestrate motor planning.

Behavioral measurements of motor readiness in mice.

Motor planning facilitates rapid and precise execution of volitional movements. Although motor planning has been classically studied in humans and monkeys, the mouse has become an increasingly popular model system to study neural mechanisms of motor planning. It remains yet untested whether mice and primates share common behavioral features of motor planning. We combined videography and a delayed response task paradigm in an autonomous behavioral system to measure motor planning in non-body-restrained mice. Motor planning resulted in both reaction time (RT) savings and increased movement accuracy, replicating classic effects in primates. We found that motor planning was reflected in task-relevant body features. Both the specific actions prepared and the degree of motor readiness could be read out online during motor planning. The online readout further revealed behavioral evidence of simultaneous preparation for multiple actions under uncertain conditions. These results validate the mouse as a model to study motor planning, demonstrate body feature movements as a powerful real-time readout of motor readiness, and offer behavioral evidence that motor planning can be a parallel process that permits rapid selection of multiple prepared actions.

Refinement of nanoporous copper by dealloying the Al-Cu alloy in NaOH solution containing sodium dodecyl sulfate.

This work reports the refinement of nanoporous copper (NPC) ligaments by introducing the sodium dodecyl sulfate (SDS) surfactant in the dealloying process. The Al80Cu20 (at%) alloy precursor is chemically dealloyed in a mixed solution of NaOH and SDS surfactant, producing NPC with a hierarchical microstructure. Micron-scaled skeletons that build up higher level networks consist of geometrically similar nano-scaled bi-continuous ligament-pore networks at the lower level. It has been found that the size of the ligaments in the lower level networks reduces from ∼32 nm to ∼24 nm with increasing SDS concentration to 1 mM. Further increasing the SDS concentration to 5 mM only leads to a slight ligament size decrease to ∼21 nm. Remarkably, nano-sized cones are formed on the lower level network surface in the dealloying solution containing 1 mM SDS, and the cone number greatly rises when the SDS concentration increases to 5 mM. The surface diffusivity of Cu adatoms is evaluated based on the experimental data, and the refinement of the ligament as well as the formation of cones are associated with the decreased surface diffusivity and the retarded Cu adatom motions with the addition of SDS. Quantum chemical calculations and molecular dynamics simulations are performed to model the adsorption behavior of SDS. It has been found that the SDS-substrate interaction increases with the number of SDS molecules before SDS reaches saturation.

Not everything, not everywhere, not all at once: a study of brain-wide encoding of movement.

Activity related to movement is found throughout sensory and motor regions of the brain. However, it remains unclear how movement-related activity is distributed across the brain and whether systematic differences exist between brain areas. Here, we analyzed movement related activity in brain-wide recordings containing more than 50,000 neurons in mice performing a decision-making task. Using multiple techniques, from markers to deep neural networks, we find that movement-related signals were pervasive across the brain, but systematically differed across areas. Movement-related activity was stronger in areas closer to the motor or sensory periphery. Delineating activity in terms of sensory- and motor-related components revealed finer scale structures of their encodings within brain areas. We further identified activity modulation that correlates with decision-making and uninstructed movement. Our work charts out a largescale map of movement encoding and provides a roadmap for dissecting different forms of movement and decision-making related encoding across multi-regional neural circuits.

Superior colliculus cell types bidirectionally modulate choice activity in frontal cortex.

Action selection occurs through competition between potential choice options. Neural correlates of choice competition are observed across frontal cortex and downstream superior colliculus (SC) during decision-making, yet how these regions interact to mediate choice competition remains unresolved. Here we report that cell types within SC can bidirectionally modulate choice competition and drive choice activity in frontal cortex. In the mouse, topographically matched regions of frontal cortex and SC formed a descending motor pathway for directional licking and a re-entrant loop via the thalamus. During decision-making, distinct neuronal populations in both frontal cortex and SC encoded opposing lick directions and exhibited push-pull dynamics. SC GABAergic neurons encoded ipsilateral choice and glutamatergic neurons encoded contralateral choice, and activating or suppressing these cell types could bidirectionally drive push-pull choice activity in frontal cortex. These results thus identify SC as a major locus to modulate choice competition within the broader action selection network.

Novel compound heterozygous mutations in TELO2 in an infant with You-Hoover-Fong syndrome: A case report and literature review.

We report here the clinical diagnosis and treatment and genetic mutations of an infant with You-Hoover-Fong syndrome (YHFS). The relevant literature review was conducted. A female infant aged 17 months was admitted to Nanhai Affiliated Maternity and Children's Hospital of Guangzhou University of Chinese Medicine due to "global development delay complicated with postnatal growth retardation for more than 1 year." The infant was diagnosed with YHFS due to the onset of extremely severe mental retardation, microcephaly, abnormal hearing, severe protein-energy malnutrition, congenital cataract, cleft palate (I°), congenital atrial septal defect, brain atrophy, hydrocephalus, and brain hypoplasia. The whole exon sequencing revealed two compound heterozygous mutations, including a likely pathogenic TELO2 variant, c.2245A > T (p.K749X) from her mother and an uncertain variant, c.2299C > T (p.R767C) from her father, validated by Sanger sequencing. After bilateral cataract surgery, the infant obtained better visual acuity and showed more responses and interactions with her parents. Diagnosis and treatment of this case prompt that these TELO2 variants have not been reported, deepening the understanding of the molecular and genetic mechanism of YHFS in clinical practice.

Characterization of Two Novel Rumen-Derived Exo-Polygalacturonases: Catalysis and Molecular Simulations.

Pectinases are a series of enzymes that degrade pectin and have been used extensively in the food, feed, and textile industries. The ruminant animal microbiome is an excellent source for mining novel pectinases. Two polygalacturonase genes, IDSPga28-4 and IDSPga28-16, from rumen fluid cDNA, were cloned and heterologously expressed. Recombinant IDSPGA28-4 and IDSPGA28-16 were stable from pH 4.0 to 6.0, with activities of 31.2 ± 1.5 and 330.4 ± 12.4 U/mg, respectively, against polygalacturonic acid. Hydrolysis product analysis and molecular dynamics simulation revealed that IDSPGA28-4 was a typical processive exo-polygalacturonase and cleaved galacturonic acid monomers from polygalacturonic acid. IDSPGA28-16 cleaved galacturonic acid only from substrates with a degree of polymerization greater than two, suggesting a unique mode of action. IDSPGA28-4 increased the light transmittance of grape juice from 1.6 to 36.3%, and IDSPGA28-16 increased the light transmittance of apple juice from 1.9 to 60.6%, indicating potential application in the beverage industry, particularly for fruit juice clarification.

Behavioral measurements of motor readiness in mice.

Motor planning facilitates rapid and precise execution of volitional movements. Although motor planning has been classically studied in humans and monkeys, the mouse has become an increasingly popular model system to study neural mechanisms of motor planning. It remains yet untested whether mice and primates share common behavioral features of motor planning. We combined videography and a delayed response task paradigm in an autonomous behavioral system to measure motor planning in non-body- restrained mice. Motor planning resulted in both reaction time savings and increased movement accuracy, replicating classic effects in primates. We found that motor planning was reflected in task-relevant body features. Both the specific actions prepared and the degree of motor readiness could be read out online during motor planning. The online readout further revealed behavioral evidence of simultaneous preparation for multiple actions under uncertain conditions. These results validate the mouse as a model to study motor planning, demonstrate body feature movements as a powerful real-time readout of motor readiness, and offer behavioral evidence that motor planning can be a parallel process that permits rapid selection of multiple prepared actions.

Hierarchical neural network with efficient selection inference.

The image classification precision is vastly enhanced with the growing complexity of convolutional neural network (CNN) structures. However, the uneven visual separability between categories leads to various difficulties in classification. The hierarchical structure of categories can be leveraged to deal with it, but a few CNNs pay attention to the character of data. Besides, a network model with a hierarchical structure is promising to extract more specific features from the data than current CNNs, since, for the latter, all categories have the same fixed number of layers for feed-forward computation. In this paper, we propose to use category hierarchies to integrate ResNet-style modules to form a hierarchical network model in a top-down manner. To extract abundant discriminative features and improve the computation efficiency, we adopt residual block selection based on coarse categories to allocate different computation paths. Each residual block works as a switch to determine the JUMP or JOIN mode for an individual coarse category. Interestingly, since some categories need less feed-forward computation than others by jumping layers, the average inference time cost is reduced. Extensive experiments show that our hierarchical network achieves higher prediction accuracy with similar FLOPs on CIFAR-10 and CIFAR-100, SVHM, and Tiny-ImageNet datasets compared to original residual networks and other existing selection inference methods.

Cloning, expression, and characterization of two pectate lyases isolated from the sheep rumen microbiome.

Pectate lyases (Pels) have a vital function in degradation of the primary plant cell wall and the middle lamella and have been widely used in the industry. In this study, two pectate lyase genes, IDSPel16 and IDSPel17, were cloned from a sheep rumen microbiome. The recombinant enzymes were expressed in Escherichia coli and functionally characterized. Both IDSPel16 and IDSPel17 proteins had an optimal temperature of 60 ℃, and an optimal pH of 10.0. IDSPel16 was relatively stable below 60 °C, maintaining 77.51% residual activity after preincubation at 60 °C for 1 h, whereas IDSPel17 denatured rapidly at 60 °C. IDSPel16 was relatively stable between pH 6.0 and 12.0, after pretreatment for 1 h, retaining over 60% residual activity. IDSPel16 had high activity towards polygalacturonic acid, with a Vmax of 942.90 ± 68.11, whereas IDSPel17 had a Vmax of only 28.19 ± 2.23 µmol/min/mg. Reaction product analyses revealed that IDSPel17 liberated unsaturated digalacturonate (uG2) and unsaturated trigalacturonate (uG3) from the substrate, indicating a typical endo-acting pectate lyase (EC 4.2.2.2). In contrast, IDSPel16 initially generated unsaturated oligogalacturonic acids, then converted these intermediates into uG2 and unsaturated galacturonic acid (uG1) as end products, a unique depolymerization profile among Pels. To the best of our knowledge, the IDSPel16 discovered with both endo-Pel (EC 4.2.2.2) and exo-Pel (EC 4.2.2.9) activities. These two pectate lyases, particularly the relatively thermo- and pH-stable IDSPel16, will be of interest for potential application in the textile, food, and feed industries. KEY POINTS: • Two novel pectate lyase genes, IDSPel16 and IDSPel17, were isolated and characterized from the sheep rumen microbiome. • Both IDSPel16 and IDSPel17 are alkaline pectate lyases, releasing unsaturated digalacturonate and unsaturated trigalacturonate from polygalacturonic acid. • IDSPel16, a bifunctional pectate lyase with endo-Pel (EC 4.2.2.2) and exo-Pel (EC 4.2.2.9) activities, could be a potential candidate for industrial application.

[A case of mental retardation caused by a frameshift variant of SYNGAP1 gene].

OBJECTIVE: To explore the genetic basis for a child with mental retardation. METHODS: Whole exome sequencing was carried out for the child. Candidate variant was screened based on his clinical features and verified by Sanger sequencing. RESULTS: The child was found to harbor a c.995_1002delAGACAAAA(p.Asp332AlafsTer84) frameshift variant in the SYNGAP1 gene. Bioinformatic analysis suggested it to be pathogenic. The same variant was not detected in either parent. CONCLUSION: The c.995_1002delAGACAAAA(p.Asp332AlafsTer84) frameshift variant of the SYNGAP1 gene probably underlay the mental retardation in this child. Above finding has expanded the spectrum of SYNGAP1 gene variants and provided a basis for the diagnosis and treatment for this child.

Immunomodulatory Blood-Derived Hybrid Hydrogels as Multichannel Microenvironment Modulators for Augmented Bone Regeneration.

Autologous blood-derived protein hydrogels have shown great promise in the field of personalized regenerative medicine. However, the inhospitable regenerative microenvironments, especially the unfavorable immune microenvironment, are closely associated with their limited tissue-healing outcomes. Herein, novel immunomodulatory blood-derived hybrid hydrogels (PnP-iPRF) are rationally designed and constructed for enhanced bone regeneration via multichannel regulation of the osteogenic microenvironment. Such double-network hybrid hydrogels are composed of clinically approved injectable platelet-rich fibrin (i-PRF) and polycaprolactone/hydroxyapatite composite nanofibers by using enriched polydopamine (PDA) as the anchor. The polycaprolactone component in PnP-iPRF provides a reinforced structure to stimulate osteoblast differentiation in a proper biomechanical microenvironment. Most importantly, the versatile PDA component in PnP-iPRF can not only offer high adhesion capacity to the growth factors of i-PRF and create a suitable biochemical microenvironment for sustained osteogenesis but also reprogram the osteoimmune microenvironment via the induction of M2 macrophage polarization to promote bone healing. The present study will provide a new paradigm to realize enhanced osteogenic efficacy by multichannel microenvironment regulations and give new insights into engineering high-efficacy i-PRF hydrogels for regenerative medicine.

A multifunctional black phosphorus-based adhesive patch intrinsically induces partial EMT for effective burn wound healing.

The ultimate goal of cutaneous wound healing is to reform a stratified epithelium to restore the normal epidermal barrier, which involves the epithelial-to-mesenchymal transition (EMT) process. However, healing strategies based on EMT induction are immature and ambiguous to date. Excessive induction of EMT may cause fibrosis, hypertrophic scarring, and increased risk of malignancy. Here, we present a new EMT-inducing strategy for eliciting partial EMT to facilitate proper epithelial cell migration. The new EMT-inducing system integrates black phosphorus nanosheets (BPNSs), catechol-modified chitosan (CA-CS), and oxidized dextran (Odex) to engineer an adhesive hydrogel patch (C&BP-Patch) with remarkable efficacy on infectious burn wound healing. The C&BP-Patch can orchestrate key early skin wound healing processes including hemostasis, inflammation, and proliferation, which enable fast partial EMT induction to restore an intact epithelial barrier. The C&BP-Patch acts initially as a high-performance bio-sealant to create a moist and stable microenvironment for EMT. Moreover, the photothermal effects of the C&BP-Patch can eliminate bacteria, accelerate microcirculation and reduce inflammation to maintain a proper EMT. Most importantly, the BPNSs can intrinsically induce partial EMT of epithelial cells via a Snail1-mediated signaling pathway. Therefore, our study proposes a new strategy for effective infectious burn wound healing based on inducing partial EMT.

Thalamus-driven functional populations in frontal cortex support decision-making.

Neurons in frontal cortex exhibit diverse selectivity representing sensory, motor and cognitive variables during decision-making. The neural circuit basis for this complex selectivity remains unclear. We examined activity mediating a tactile decision in mouse anterior lateral motor cortex in relation to the underlying circuits. Contrary to the notion of randomly mixed selectivity, an analysis of 20,000 neurons revealed organized activity coding behavior. Individual neurons exhibited prototypical response profiles that were repeatable across mice. Stimulus, choice and action were coded nonrandomly by distinct neuronal populations that could be delineated by their response profiles. We related distinct selectivity to long-range inputs from somatosensory cortex, contralateral anterior lateral motor cortex and thalamus. Each input connects to all functional populations but with differing strength. Task selectivity was more strongly dependent on thalamic inputs than cortico-cortical inputs. Our results suggest that the thalamus drives subnetworks within frontal cortex coding distinct features of decision-making.

Molecular subtype identification and predictive power of N6-methyladenosine regulator in unexplained recurrent pregnancy loss.

The etiology of recurrent pregnancy loss (RPL) is complicated and effective clinical preventive measures are lacking. Identifying biomarkers for RPL has been challenging, and to date, little is known about the role of N6-methyladenosine (m6A) regulators in RPL. Expression data for m6A regulators in 29 patients with RPL and 29 healthy controls were downloaded from the Gene Expression Omnibus (GEO) database. To establish a diagnostic model for unexplained RPL, differential gene expression analysis was conducting for 36 m6A regulators using least absolute shrinkage and selection operator (LASSO) regression. Unsupervised cluster analysis was conducted on hub genes, and probable mechanisms were explored using gene set enrichment analysis (GSEA) and gene ontology (GO) analysis. Correlations between m6A-related differentially expressed genes and immune infiltration were analyzed using single-sample GSEA. A total of 18 m6A regulators showed significant differences in expression in RPL: 10 were upregulated and eight were downregulated. Fifteen m6A regulators were integrated and used to construct a diagnostic model for RPL that had good predictive efficiency and robustness in differentiating RPL from control samples, with an overall area under the curve (AUC) value of 0.994. Crosstalk was identified between 10 hub genes, miRNAs, and transcription factors (TFs). For example, YTHDF2 was targeted by mir-1-3p and interacted with embryonic development-related TFs such as FOXA1 and GATA2. YTHDF2 was also positively correlated with METTL14 (r = 0.5983, p < 0.001). Two RPL subtypes (Cluster-1 and Cluster-2) with distinct hub gene signatures were identified. GSEA and GO analysis revealed that the differentially expressed genes were mainly associated with immune processes and cell cycle signaling pathway (normalized enrichment score, NES = -1.626, p < 0.001). Immune infiltration was significantly higher in Cluster-1 than in Cluster-2 (p < 0.01). In conclusion, we demonstrated that m6A modification plays a critical role in RPL. We also developed and validated a diagnostic model for RPL prediction based on m6A regulators. Finally, we identified two distinct RPL subtypes with different biological processes and immune statuses.