Search

Based on the action of the mapping class group on the space of measured foliations, we construct a new boundary of the mapping class group and study the structure of this boundary. As an application, for any point in Teichmuller space, we consider the orbit of this point under the action of the mapping class group and describe the closure of this orbit in the Thurston compactification and the Gardiner-Masur compactification of Teichmuller space. We also construct some new points in the Gardiner-Masur boundary of Teichmuller space., Comment: To appear in Acta Mathematica Sinica, English Series

Additional file 1: Figure S1. LSP1 immunohistochemical staining controls for Fig. 2A, B. Figure S2. LSP1, MPO, and Gr1 immunofluorescent staining controls for Fig. 2C and Fig. 3. Figure S3. Cytospun BAL cells were stained with Hemacolor stain kit

Additional file 2. Full length gels and blots with membrane edges.

In order to combat severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic in the world, we formulate S1 subunit of the virus with two types of adjuvants, amphiphilic adjuvant monophosphoryl lipid A (MPLA) for Toll-like receptor 4 (TLR4) and CpG ODN for TLR9, into cationic multifunctional liposomes to produce a potent, safer, and translatable nanovaccine. The results show that the nanovaccine can efficiently elicit humoral immune response in mice. The sera from the vaccinated mice significantly inhibit SARS-CoV-2 to infect Vero cells. Moreover, relatively to the free S1 with traditional Alum adjuvant, the nanovaccine can elicit strong T cell immunity by activating both CD4+ and CD8+ cells, which may play critical roles in eliminating viral load in patients. Most importantly, the nanovaccine can elicit strong IgA antibody, providing potential mucosal protection to host. Altogether, this study offers a translatable design for a potent subunit SARS-CoV-2 nanovaccine.

In our life, we often need to estimate the accuracy of sports pose, which usually costs a lot of time and human resources. To solve the problem, we propose a LSTM-Attention model. In spatial dimension, we use the two-branch multi-stage CNN to extract human joints as features, which not only guarantees the real-time performance, but also ensures the accuracy. For the time dimension, the extracted joint features sequence is input into the LSTM-Attention model for training. In order to verify the effectiveness of our proposed method, we collected data for processing and trained with the proposed model. The experimental results show that our method has a high performance.

The avian EB66® cell line, derived from duck embryonic stem cells, has been widely used for producing human and animal therapeutic proteins and vaccines. In current study we evaluated the potential use of EB66® cell line in a cell culture-derived duck Tembusu virus (DTMUV) vaccine development. After optimizing the growth conditions of DTMUV HB strain in EB66® cells, we successfully generated three batches of viruses with ELD50 titres of 105.9/0.1 ml, 105.3/0.1 ml and 105.5/0.1 ml, respectively, for using in the preparation of inactivated vaccines. The immunogenicity and protective efficacy of these EB66® cells-derived inactivated vaccines were examined in ducks. Results indicated that all three batches of vaccines induced haemagglutination-inhibition (HI) antibody response in immunized birds at 2 weeks after a single immunization. Immunized ducks and ducklings were protected against a virulent challenge at 4 weeks after a booster immunization. The duration of immunity was for 3–4 months after a booster immunization. These results demonstrated the feasibility of using EB66® cell line to grow up DTMUV for vaccine preparation. RESEARCH HIGHLIGHTSDuck Tembusu virus can be propagated in EB66® cells.EB66® cell-derived inactivated DTMUV vaccines are immunogenic and can provide protection against a virulent challenge.A long-lasting immunity is induced after a booster immunization. Duck Tembusu virus can be propagated in EB66® cells. EB66® cell-derived inactivated DTMUV vaccines are immunogenic and can provide protection against a virulent challenge. A long-lasting immunity is induced after a booster immunization.

The memristor device based on graphene/Nb-doped SrTiO3 heterojunction was fabricated. The device exhibits obvious resistive switching between high resistance state and low resistance state. By using the graphene as the transparent electrode, remarkable photocurrent and photovoltage was achieved in the heterojunction. The photovoltage shows a clear dependence on the resistance state, whereas the photocurrent keeps unchanged. Based on these features, a new non-destructive photovoltaic reading method of the memristor was demonstrated. The photovoltaic reading method is free of external driving voltage, thereby avoiding the perturbation of the stored data. Moreover, the photovoltage output signal instead of the current is more compatible with the IC technologies.

Dissolution of the exposed sphalerite (marmatite) in abandoned mining sites and tailings may exacerbate acid and metalliferous drainage (AMD) hazards. Cupric ions are inevitable ions in AMD systems but its action mechanism on the dissolution of sphalerite is still unclear. In this work, the possible phase transition from sphalerite to chalcopyrite is firstly discovered in acidic cupric ions solution according to the results of Raman and (synchrotron radiation-based) X-ray (micro-) diffractometer spectra, which should be an important reason that mediates the dissolution of sphalerite. Results of DFT calculations reveal the underlying mechanism that Cu2+ can selectively replace zinc in marmatite lattices and further diffuse into the matrix. Additionally, a strong correlation between the cupric ion consumption with the pH value variation is discussed and the effects of the formed new phase on the dissolution kinetics of marmatite were researched. According to this work, the action mechanism of cupric ions on sphalerite dissolution in acidic environments is furtherly clarified.

In order to improve the structural strength of the turbine rotor, the dynamic circuit and driving mechanical model of the turbine rotor are established under the thermal cycle fatigue load. The numerical simulation principle of the stress yield evolution process of the rotor structure is analysed. The closed vector method and Lagrange equation are used to establish the mechanical analysis and yield stress model, and the mechanical principle of the thermal cycle fatigue load of the rotor is analysed. On this basis, the yield stress evaluation model is established to realise the material structure evolution analysis. The results show that the method can provide higher accuracy and lower yield stress estimation results, with an average error of 15.74% lower than that of the traditional method, which shows that the method can improve the anti-interference ability of turbine rotor drive and the stress intensity of rotor material structure, with good convergence.

Accurate demand forecasting is very important for power system operation and planning. In order to forecast energy demand more accurately, a short-term power load forecasting model based on Bayesian neural network is proposed. Firstly, Pearson coefficient is used to analyze the influence of distributed energy on the load of the distribution network, and Bayesian neural network (BNN) is used to make predictions. Then, particle swarm optimization (PSO) is used to optimize the weights and thresholds of the neural network in order to obtain the best prediction effect. The experimental results show that the Pearson coefficient can quantitatively analyze the impact of distributed energy. The prediction accuracy of BNN-PSO based prediction model is significantly higher than that of other conventional methods, and the proposed prediction model has higher prediction accuracy.

The large-scale access of distributed generation has a great impact on load forecasting in substation-area, which means the load curve in the substation-area cannot reflect the real load of users. Firstly, this paper considers the impact of distributed generation on the load curve, and proposes a load forecasting model based on data cleaning and deep learning in the substation-area. Secondly, considering the data missing during communication and transmission, the KNN algorithm is adopted to complete the missing data before the data input. And then, Pearson correlation coefficients are used for the correlation analysis of the input factors related to distributed generation, and the data is trained through Long-short Term Memory in deep learning. Finally, verified by load data of some substation-area in Jiangsu Province, the prediction model established in this paper has good prediction accuracy and stability.

Neural networks are a powerful class of nonlinear functions that can be trained end-to-end on various applications. While the over-parametrization nature in many neural networks renders the ability to fit complex functions and the strong representation power to handle challenging tasks, it also leads to highly correlated neurons that can hurt the generalization ability and incur unnecessary computation cost. As a result, how to regularize the network to avoid undesired representation redundancy becomes an important issue. To this end, we draw inspiration from a well-known problem in physics -- Thomson problem, where one seeks to find a state that distributes N electrons on a unit sphere as evenly as possible with minimum potential energy. In light of this intuition, we reduce the redundancy regularization problem to generic energy minimization, and propose a minimum hyperspherical energy (MHE) objective as generic regularization for neural networks. We also propose a few novel variants of MHE, and provide some insights from a theoretical point of view. Finally, we apply neural networks with MHE regularization to several challenging tasks. Extensive experiments demonstrate the effectiveness of our intuition, by showing the superior performance with MHE regularization., Comment: NeurIPS 2018

Steam generator overfill due to steam generator tube rupture have been analyzed for CAP1400 plant, in which two cases were simulated, including pressurizer heater switching off available and unavailable under low pressurizer water level. The results indicate that it has a certain margin to steam generator (SG) overfill for the ruptured steam generator under steam generator tube rupture (SGTR) accident. Analysis results have also shown that the SG overfill margin will be slightly decreased in case of pressurizer heater swiching off unavailable, however, the impact is not significant.

We prove that the Teichm��ller space of surfaces with given boundary lengths equipped with the arc metric (resp. the Teichm��ller metric) is almost isometric to the Teichm��ller space of punctured surfaces equipped with the Thurston metric (resp. the Teichm��ller metric)., 20 pages, 5 figures. All comments are welcome!

A transducer with a fixed aperture is proposed as a virtual array detector to collect data. The transducer only needs to rotationally scan the sample 180 degrees. In the experiment, a Q-switched Nd: YAG laser operating at 532 nm was used as the light source, with a pulse width of 7 ns and a repetition frequency of 10 Hz. An unfocused transducer with a central frequency of 2.25 MHz and a diameter of 6 mm was applied to detecting photoacoustic signals. And the image was reconstructed by processing the received photoacoustic signals with the inverse Radon transform. An in vitro claw of a rat was used as the imaging object in the experiment and the data were obtained by rotating the transducer 180 degrees with a step of 2 degrees. The reconstructed image from the inverse Radon transform clearly displays the bones within the claw. The results indicate that the inverse Radon transform can be used for photoacoustic reconstruction and limited-view photoacoustic imaging is achieved with this method.

Recently, keyword search on XML data has received much attention. Existing XML keyword search algorithms are all based on Dewey labeling, and this method in the calculation of the common ancestor would suffer from the CAR (common-ancestor-repetition) problem. In this paper, we propose a novel index based on interval encoding, namely IS-Index. We then design an efficient algorithm based on this structure, namely Interval-Search algorithm. Firstly, the algorithm calculates the minimum common ancestors by the range value between nodes that contain keywords. Then it traverses the IS-Index and computes common ancestor nodes to get the candidate SLCAs. Finally, it filters the candidate SLCAs to get SLCA nodes. The experimental results show that our method is an efficient algorithm for finding SLCA nodes.

Let $X_{0}$ be a complete hyperbolic surface of infinite type with geodesic boundary which admits a countable pair of pants decomposition. As an application of the Basmajian identity for complete bordered hyperbolic surfaces of infinite type with limit sets of 1-dimensional measure zero, we define an asymmetric metric (which is called arc metric) on the quasiconformal Teichm��ller space $\mathcal{T}(X_{0})$ provided that $X_{0}$ satisfies a geometric condition. Furthermore, we construct several examples of hyperbolic surfaces of infinite type satisfying the geometric condition and discuss the relation between the Shiga's condition and the geometric condition., 31 pages, 20 figures

An unprecedented 3D organic–inorganic hybrid compound [Cu7(bipy)7Cl2(AsW12O40)]·2H2O (1) has been hydrothermally synthesized and characterized by elemental analyses, IR and single-crystal X-ray diffraction. Compound 1 crystallizes in the triclinic system, space group P-1 with a = 12.292(3) A, b = 13.115(3) A, c = 15.346(3) A, α = 99.11(3)°, β = 91.35(3)°, γ = 104.34(3)°, V = 2361.5(8) A3. X-ray diffraction reveals that compound 1 is the first 3D organic–inorganic hybrid based on the arsenotungstate Keggin POMs, in which the big 12-membered rings and the small four-membered ring dimers construct a 2D multi-nuclear Cu(I) cluster. The result again shows that the formation of the cyclic structure in the present work is attributed to the role of the polyoxoanion. Additionally, changing Cu metal ion into Ag, compound 2 (bipy)4[AsW12O40]·2H2O has been obtained and exhibits a supramolecular structure.

Frequent item sets mining algorithms in uncertain data streams almost base on the expected frequent item sets. Compared to probabilistic frequent item sets, it can't reflect the confidence of item sets. We propose the algorithm based on probabilistic frequent item sets mining in uncertain data streams. The algorithm processes one basic sliding window every time, and the mining results are stored in the Probabilistic Frequent Tree. When the window sliding, it dynamically updates Probabilistic Frequent Tree to delete old data and add new data. Theoretical analysis and experiments show that the algorithm is effective.

The arc metric is an asymmetric metric on the Teichm{��}ller space T(S) of a surface S with nonempty boundary. In this paper we study the relation between Thurston's compactification and the horofunction compactification of T(S) endowed with the arc metric. We prove that there is a natural homeomorphism between the two compactifications.

The Peng-Robinson equation of state PR EOS can accurately describe both the liquid and vapour phase. We present several applications of this equation of state with the aid of Mathcad, including estimation of pure component properties, computation of vapour-liquid equilibrium VLE and isobaric diagram for a binary mixture. In addition, high-pressure chemical-equilibrium calculations are performed. An example of an adiabatic flash computation is also presented.

Given a surface of infinite topological type, there are several Teichm��ller spaces associated with it, depending on the basepoint and on the point of view that one uses to compare different complex structures. This paper is about the comparison between the quasiconformal Teichm��ller space and the length-spectrum Teichm��ller space. We work under this hypothesis that the basepoint is upper-bounded and admits short interior curves. There is a natural inclusion of the quasiconformal space in the length-spectrum space. We prove that, under the above hypothesis, the image of this inclusion is nowhere dense in the length-spectrum space. As a corollary we find an explicit description of the length-spectrum Teichm��ller space in terms of Fenchel-Nielsen coordinates and we prove that the length-spectrum Teichm��ller space is path-connected.

We show that the length spectrum metric on Teichm��ller spaces of surfaces of infinite topological type is complete. We also give related results and examples that compare the length spectrum Teichm��ller space with quasiconformal and the Fenchel-Nielsen Teichm��ller spaces on such surfaces

We prove an analogue of Farb-Masur's theorem that the length-spectra metric on moduli space is "almost isometric" to a simple model $\mathcal {V}(S)$ which is induced by the cone metric over the complex of curves. As an application, we know that the Teichm��ller metric and the length-spectra metric are "almost isometric" on moduli space, while they are not even quasi-isometric on Teichm��ller space.

There are several Teichm��ller spaces associated to a surface of infinite topological type, after the choice of a particular basepoint (a complex or a hyperbolic structure on the surface). These spaces include the quasiconformal Teichm��ller space, the length spectrum Teichm��ller space, the Fenchel-Nielsen Teichm��ller space, and there are others. In general, these spaces are set-theoretically different. An important question is therefore to understand relations between these spaces. Each of these spaces is equipped with its own metric, and under some hypotheses, there are inclusions between these spaces. In this paper, we obtain local metric comparison results on these inclusions, namely, we show that the inclusions are locally bi-Lipschitz under certain hypotheses. To obtain these results, we use some hyperbolic geometry estimates that give new results also for surfaces of finite type. We recall that in the case of a surface of finite type, all these Teichm��ller spaces coincide setwise. In the case of a surface of finite type with no boundary components (and possibly with punctures), we show that the restriction of the identity map to any thick part of Teichm��ller space is globally bi-Lipschitz with respect to the length spectrum metric and the classical Teichm��ller metric on the domain and on the range respectively. In the case of a surface of finite type with punctures and boundary components, there is a metric on the Teichm��ller space which we call the arc metric, whose definition is analogous to the length spectrum metric, but which uses lengths of geodesic arcs instead of lengths of closed geodesics. We show that the restriction of the identity map restricted to any "relative thick" part of Teichm��ller space is globally bi-Lipschitz, with respect to any of the three metrics: the length spectrum metric, the Teichm��ller metric and the arc metric on the domain and on the range.

We show that the horofunction compactification of Teichm��ller space with the Teichm��ller metric is homeomorphic to the Gardiner-Masur compactification., Final version, the proof of Proposition 5.1 is improved. To appear in Handbook of Teichm\"uller Theory, Vol. 4, A. Papadopoulos (ed.), EMS publishing house, Z\"urich 2013

We introduce Fenchel-Nielsen coordinates on Teicm��ller spaces of surfaces of infinite type. The definition is relative to a given pair of pants decomposition of the surface. We start by establishing conditions under which any pair of pants decomposition on a hyperbolic surface of infinite type can be turned into a geometric decomposition, that is, a decomposition into hyperbolic pairs of pants. This is expressed in terms of a condition we introduce and which we call Nielsen convexity. This condition is related to Nielsen cores of Fuchsian groups. We use this to define the Fenchel-Nielsen Teichm��ller space associated to a geometric pair of pants decomposition. We study a metric on such a Teichm��ller space, and we compare it to the quasiconformal Teichm��ller space, equipped with the Teichm��ller metric. We study conditions under which there is an equality between these Teichm��ller spaces and we study topological and metric properties of the identity map when this map exists.

This paper is concerned with the exponential stability of the zero solution of a quasi-linear differential equation in a Banach space, it is proved that the quasi-linear non-autonomous differential equation is exponentially stable if and only if the associated linear non-autonomous differential equation is exponentially stable.

Unlike the case of surfaces of topologically finite type, there are several different Teichm��ller spaces that are associated to a surface of topological infinite type. These Teichm��ller spaces first depend (set-theoretically) on whether we work in the hyperbolic category or in the conformal category. They also depend, given the choice of a point of view (hyperbolic or conformal), on the choice of a distance function on Teichm��ller space. Examples of distance functions that appear naturally in the hyperbolic setting are the length spectrum distance and the bi-Lipschitz distance, and there are other useful distance functions. The Teichm��ller spaces also depend on the choice of a basepoint. The aim of this paper is to present some examples, results and questions on the Teichm��ller theory of surfaces of infinite topological type that do not appear in the setting the Teichm��ller theory of surfaces of finite type. In particular, we point out relations and differences between the various Teichm��ller spaces associated to a given surface of topological infinite type., The revised version will appear in the Transactions of the AMS

Recently nano-materials are widely used but they have shown contrasting effects on human and plant life. Keeping in view the contrasting results, the present study has evaluated plant growth response, antioxidant system activity and photosynthetic apparatus physiological and ultrastructural changes in Brassica napus L. plants grown under a wide range (0, 500, 2500, 4000 mg/l) of nano-TiO2 in a pot experiment. Nano-TiO2 has significantly improved the morphological and physiological indices of oilseed rape plants under our experimental conditions. All the parameters i-e morphological (root length, plant height, fresh biomass), physiological (photosynthetic gas exchange, chlorophyll content, nitrate reductase activity) and antioxidant system (Superoxide dismutase, SOD; Guaiacol peroxidase, POD; Catalase, CAT) recorded have shown improvement in their performance by following nano-TiO2 dose-dependent manner. No significant chloroplast ultra-structural changes were observed. Transmission electron microscopic images have shown that intact & typical grana and stroma thylakoid membranes were in the chloroplast, which suggest that nano-TiO2 has not induced the stressful environment within chloroplast. Finally, it is suggested that, nano-TiO2 have growth promoting effect on oilseed rape plants.

According to the environment and work styles, the underwater monitoring devices can be divided into three type, fixed monitoring device, the floating monitoring device and mobile monitoring device. Floating monitoring platform is a new type of monitoring device of the floating monitoring device, which is mainly used for underwater video monitoring of pool. as the floating platform monitoring in water motion and hydrostatic characteristics are closely related, the influent of counterweight, weight distance and floating body diameter of the structure parameters on its hydrostatic characteristics and the natural periods for roll, pitch and heave should be considered, In this work, the floating body diameter influent most, the counterweight followed, and the structure parameters have different influence on the natural periods for roll, pitch and heave , it need to be analyzed according to the concrete structure parameters.

We have made several reports on the signal transduction mechanism that emodin enhance the calcium concentrations of smooth muscle cells (SMCs) in the physiological condition by inositol [ 1 , 4 , 5 ]-friphosphate (IP 3 ). The observation that IP3 concentrations in SMCs were decreased in multiple organ dysfunction syndrome (MODS) prompted us to ask whether emodin can activate SMCs to contract by way of elevating [Ca 2+ ] and thus modulating the critical Ca 2+ signal transduction pathways involved in the contraction of the SMCs in the pathological setting of MODS. To test this hypothesis, we used the rat model of MODS to explore the potential roles of emodin in Ca 2+ signal transduction in the SMCs of colon in rats. ML-7 [an inhibitor of myosin light-chain kinase (MLCK)] and Calphostin C [an inhibitor of protein kinase C (PKC)] were used to observe the influence of emodin on the muscle strips and SMCs in rats after MODS. Nifedipine (an antagonist of voltage-gated Ca 2+ channel), EGTA (removal of extracellular Ca 2+ ), heparine (a specific IP3 receptor antagonist), and ryanodine were used to probe the potential mechanisms involved in emodin-mediated elevation of the global cytoplasmic Ca 2+ in SMCs of colon in the rats after MODS. Our results show that emodin is capable of contract the smooth muscles of colon in rats after MODS by MLCK increasing [Ca 2+ ] of SMCs, and by PKC enhancing the calcium sensitivity of SMCs. The mechanism by which emodin triggers elevated [Ca 2+ ] of smooth muscles of colon in rats after MODS is likely to operate through IP 3 and RyR receptors in the sarcoplasm. It is hoped that deeper insights into how emodin modulates the critical calcium signaling in SMCs might lead to the potential development of emodin in the treatment of MODS.

Fixed field experimental studies are carried out on daily variations of the undisturbed community and soil respiration fluxes in different phenological phases of 2001―2002 in semiarid Aneurolepidium chinense steppe of Inner Mongolia, China using static black chamber method. Corresponding statistical analysis of the contributions of the water-heat factors (air temperature, ground temperature, surface soil water content) and ecological factors (aboveground biomass, underground biomass, litter biomass) to daily variation law of the undisturbed community and soil respiration fluxes as well as differences in daily respiration are also conducted. The results indicate that undisturbed community and soil respiration have apparent daily variation laws, daily variation patterns of respiration fluxes during different phenological phases are basically the same, and the variations of environmental factors only exert effect on CO 2 emission intensities, while the effect on daily variation pattern of grassland CO 2 emission fluxes is relatively small. The daily total respiration of the undisturbed community in different phenological phases ranges from 1.34―10.13 g·m −2 ; soil daily total respiration ranges from 0.98―5.17 g·m −2 ; both daily variations of undisturbed community and soil respiration fluxes are significantly correlated ( p p < 0.01) with air temperatures and ground surface temperatures, but the correlativity with the soil temperature at 5 and 10 cm depths is relatively weak; multiple regression analysis indicates that about 80% of the difference in daily respiration of the undisturbed community among different phenological phases is induced by the variation of the aboveground biomass, while the variations of the remaining factors can jointly explain around 20% of the daily respiration variations of the whole grassland ecosystem; about 83% of the soil daily respiration variation of the different phenological phases is caused by 0―20 cm underground biomass. Besides, surface soil water content is also an important environmental factor affecting soil daily respiration variations of the Aneurolepidium chinense steppe, but its partial correlation coefficient with soil daily respiration amount does not reach the significance level of 0.05.

Suggestions for improving the mathematical formula for determining group electronegativities.

Directional solidification experiments have been carried out to study the variation in primary and secondary arm spacings with solidification parameters in the Al-Cu system. It is found that the primary arm spacing Z1 obeys the common correlation Z1 = KG-aV-b in the high velocity regime or low temperature gradient regime, where a and b are constants, the value of K included the composition dependence, G is the temperature gradient in the liquid and V the growth rate; however, it does not obey this correlation for the low velocity regime or high temperature gradient regime but goes through a maximum or a catastrophe as a function of V or G at V = V cs k or G = kGcs, where k is the equilibrium distribution coefficient, and Vcs and Gcs are the critical velocity and temperature gradient at the limit of constitutional undercooling respectively. The initial secondary arm spacings Z20 are nearly independent of G and mainly depended on V, Z20 = 0.016 V-0.54 (mm). The secondary arm spacing Z2 tends to c oarsen with time and thus is a function of coarsening time tf, Z2 = 0.016t0.34f (mm). Theoretical analyses of the primary arm spac ing and the initial secondary arm spacing have been proposed, and the derived relationships agree reasonably well with the above experimental results.

Bone metastasis (BM) is the advanced complication of breast cancer, while bone marrow-derived mesenchymal stem cells (BMSCs) in the microenvironment unclearly contribute to cancer metastasis. This study investigated potential roles of transforming growth factor- (TGF-) α in the interaction between breast cancer and BMSCs in BM. Clinical cases of breast cancer with bone metastasis (BMBC), breast cancer without bone metastasis (Non-BM-BC), and benign fibroadenoma (Benign) were enlisted in a retrospective study. TGF-α was found obviously overexpressed in BM lesion of BMBC compared to primary lesion of both BMBC and Non-BM-BC (P<0.01), and TGF-α was higher in primary lesion of both BMBC and Non-BM-BC (P<0.01) than Benign group. Interestingly, TGF-α in nontumor tissues of both BMBC and Non-BM-BC was at a higher level than Benign group (P<0.01), and numbers of macrophages in nontumor tissues of both BMBC and Non-BM-BC (P<0.01) were higher than Benign group. Furthermore, in cultured human BMSCs, TGF-α stimulated production of procancer cytokines including IL-6, VEGF, FGF10, FGF17, and TGF-β1 in a dose-dependent manner. Thus, TGF-α in BC could potentially be an important signal of carcinogenesis and metastasis. Macrophages in the nontumor tissue of BC may not be protective but could promote cancer metastasis.