arXiv subjects

Wenjuan Liu

Publications and source records attributed to Wenjuan Liu.

17 recordsLinked to original sources

arXiv2024

DEDUCE: Multi-head attention decoupled contrastive learning to discover cancer subtypes based on multi-omics data

Background and Objective: Given the high heterogeneity and clinical diversity of cancer, substantial variations exist in multi-omics data and clinical features across different cancer subtypes. Methods: We propose a model, named DEDUCE, based on a symmetric multi-head attention encoders (SMAE), for unsupervised contrastive learning to analyze multi-omics cancer data, with the aim of identifying and characterizing cancer subtypes. This model adopts a unsupervised SMAE that can deeply extract contextual features and long-range dependencies from multi-omics data, thereby mitigating the impact of noise. Importantly, DEDUCE introduces a subtype decoupled contrastive learning method based on a multi-head attention mechanism to simultaneously learn features from multi-omics data and perform clustering for identifying cancer subtypes. Subtypes are clustered by calculating the similarity between samples in both the feature space and sample space of multi-omics data. The fundamental concept involves decoupling various attributes of multi-omics data features and learning them as contrasting terms. A contrastive loss function is constructed to quantify the disparity between positive and negative examples, and the model minimizes this difference, thereby promoting the acquisition of enhanced feature representation. Results: The DEDUCE model undergoes extensive experiments on simulated multi-omics datasets, single-cell multi-omics datasets, and cancer multi-omics datasets, outperforming 10 deep learning models. The DEDUCE model outperforms state-of-the-art methods, and ablation experiments demonstrate the effectiveness of each module in the DEDUCE model. Finally, we applied the DEDUCE model to identify six cancer subtypes of AML.

cs.LG↗

arXiv2024

SELECTOR: Heterogeneous graph network with convolutional masked autoencoder for multimodal robust prediction of cancer survival

Accurately predicting the survival rate of cancer patients is crucial for aiding clinicians in planning appropriate treatment, reducing cancer-related medical expenses, and significantly enhancing patients' quality of life. Multimodal prediction of cancer patient survival offers a more comprehensive and precise approach. However, existing methods still grapple with challenges related to missing multimodal data and information interaction within modalities. This paper introduces SELECTOR, a heterogeneous graph-aware network based on convolutional mask encoders for robust multimodal prediction of cancer patient survival. SELECTOR comprises feature edge reconstruction, convolutional mask encoder, feature cross-fusion, and multimodal survival prediction modules. Initially, we construct a multimodal heterogeneous graph and employ the meta-path method for feature edge reconstruction, ensuring comprehensive incorporation of feature information from graph edges and effective embedding of nodes. To mitigate the impact of missing features within the modality on prediction accuracy, we devised a convolutional masked autoencoder (CMAE) to process the heterogeneous graph post-feature reconstruction. Subsequently, the feature cross-fusion module facilitates communication between modalities, ensuring that output features encompass all features of the modality and relevant information from other modalities. Extensive experiments and analysis on six cancer datasets from TCGA demonstrate that our method significantly outperforms state-of-the-art methods in both modality-missing and intra-modality information-confirmed cases. Our codes are made available at https://github.com/panliangrui/Selector.

cs.CV↗

arXiv2023

ESM-NBR: fast and accurate nucleic acid-binding residue prediction via protein language model feature representation and multi-task learning

Protein-nucleic acid interactions play a very important role in a variety of biological activities. Accurate identification of nucleic acid-binding residues is a critical step in understanding the interaction mechanisms. Although many computationally based methods have been developed to predict nucleic acid-binding residues, challenges remain. In this study, a fast and accurate sequence-based method, called ESM-NBR, is proposed. In ESM-NBR, we first use the large protein language model ESM2 to extract discriminative biological properties feature representation from protein primary sequences; then, a multi-task deep learning model composed of stacked bidirectional long short-term memory (BiLSTM) and multi-layer perceptron (MLP) networks is employed to explore common and private information of DNA- and RNA-binding residues with ESM2 feature as input. Experimental results on benchmark data sets demonstrate that the prediction performance of ESM2 feature representation comprehensively outperforms evolutionary information-based hidden Markov model (HMM) features. Meanwhile, the ESM-NBR obtains the MCC values for DNA-binding residues prediction of 0.427 and 0.391 on two independent test sets, which are 18.61 and 10.45% higher than those of the second-best methods, respectively. Moreover, by completely discarding the time-cost multiple sequence alignment process, the prediction speed of ESM-NBR far exceeds that of existing methods (5.52s for a protein sequence of length 500, which is about 16 times faster than the second-fastest method). A user-friendly standalone package and the data of ESM-NBR are freely available for academic use at: https://github.com/wwzll123/ESM-NBR.

q-bio.QM↗

arXiv2023

PACS: Prediction and analysis of cancer subtypes from multi-omics data based on a multi-head attention mechanism model

Due to the high heterogeneity and clinical characteristics of cancer, there are significant differences in multi-omic data and clinical characteristics among different cancer subtypes. Therefore, accurate classification of cancer subtypes can help doctors choose the most appropriate treatment options, improve treatment outcomes, and provide more accurate patient survival predictions. In this study, we propose a supervised multi-head attention mechanism model (SMA) to classify cancer subtypes successfully. The attention mechanism and feature sharing module of the SMA model can successfully learn the global and local feature information of multi-omics data. Second, it enriches the parameters of the model by deeply fusing multi-head attention encoders from Siamese through the fusion module. Validated by extensive experiments, the SMA model achieves the highest accuracy, F1 macroscopic, F1 weighted, and accurate classification of cancer subtypes in simulated, single-cell, and cancer multiomics datasets compared to AE, CNN, and GNN-based models. Therefore, we contribute to future research on multiomics data using our attention-based approach.

q-bio.QM↗

arXiv2021

Robust Blind Source Separation by Soft Decision-Directed Non-Unitary Joint Diagonalization

Approximate joint diagonalization of a set of matrices provides a powerful framework for numerous statistical signal processing applications. For non-unitary joint diagonalization (NUJD) based on the least-squares (LS) criterion, outliers, also referred to as anomaly or discordant observations, have a negative influence on the performance, since squaring the residuals magnifies the effects of them. To solve this problem, we propose a novel cost function that incorporates the soft decision-directed scheme into the least-squares algorithm and develops an efficient algorithm. The influence of the outliers is mitigated by applying decision-directed weights which are associated with the residual error at each iterative step. Specifically, the mixing matrix is estimated by a modified stationary point method, in which the updating direction is determined based on the linear approximation to the gradient function. Simulation results demonstrate that the proposed algorithm outperforms conventional non-unitary diagonalization algorithms in terms of both convergence performance and robustness to outliers.

eess.SP↗

arXiv2021

Mrk 1239: a Type-2 Counterpart of Narrow-line Seyfert-1?

We present new spectrophotometric and spectropolarimetric observations of Mrk 1239, one of the 8 prototypes that defines type-1 narrow-line Seyfert galaxies (NLS1s). Unlike the other typical NLS1s though, a high degree of polarization ($P\sim$5.6%) and red optical-IR ($g-W_4$ = 12.35) colors suggest that Mrk 1239 is more similar to type-2 active galactic nuclei like NGC 1068. Detailed analysis of spectral energy distribution in the UV-optical-IR yields two components from the nucleus: a direct and transmitted component that is heavily obscured ($E_{B-V} \approx 1.6$), and another indirect and scattered one with mild extinction ($E_{B-V} \sim$ 0.5). Such a two-light-paths scenario is also found in previous reports based on the X-ray data. Comparison of emission lines and the detection of He\,{\footnotesize I}*$λ$10830 BAL at [-3000,-1000] km s$^{-1}$ indicates that the obscuring clouds are at physical scale between the sublimation radius and that of the narrow emission line regions. The potential existence of powerful outflows is found as both the obscurer and scatterer are outflowing. Similar to many other type-2s, jet-like structure in the radio band is found in Mrk 1239, perpendicular to the polarization angle, suggesting polar scattering. We argue that Mrk 1239 is very probably a type-2 counterpart of NLS1s. The identification of 1 out of 8 prototype NLS1s as a type-2 counterpart implies that there can be a substantial amount of analogs of Mrk 1239 misidentified as type-1s in the optical band. Properties of these misidentified objects are going to be explored in our future works.

astro-ph.GA↗

arXiv2019

A Deeply Buried Narrow-Line Seyfert 1 Nucleus Uncovered in Scattered Light

We present spectropolarimetric and spectrophotometric observations of the peculiar active galactic nucleus (AGN) SDSS J120300.19+162443.7 (hereafter J1203+1624) at $z=0.1656$. Its optical total flux spectra clearly show broad emission lines (BELs) in H$α$ and H$β$. After removal of narrow emission lines (NELs), the full width at half maximum of the lorentzian BEL is $FWHM_{\rm BEL} \approx 1,270$ km s$^{-1}$, fulfilling the conventional definition of a narrow line Seyfert 1 (NLS1) galaxy. However, its NELs are unprecedentedly strong when compared to type 1AGNs. This, together with its large MIR excess ($g-W_4 = 13.172$ mag), implies that the observer and the narrow emission line region (NELR) might see a different ionization continuum. Our optical spectropolarimetry confirms its type 2 nature by detecting a polarized blue continuum and Balmer BELs ($FWHM_{\rm Polarized~BEL} \approx 1,183$ km s$^{-1}$), with a high polarization degree of $> 20$% in the blue wing. The mass and Eddington rate of the central black hole is estimated based on both transmitted and scattered AGN radiation, which is $M_{\bullet} < 2.9 \times 10^7 M_\odot$ and $L_{\rm bol}/L_{\rm Edd} > 1.5$. Severe extinction of the AGN emission also enables a clear view of the compact host galaxy. Discovery of J1203+1624 suggests that NLS1 follows AGN unfication scheme, studying of its analogs could blaze a new trail for exploring the connection between black hole growth and star formation in the host galaxy. Interesting features of J1203+1624 like the peculiar NELs, inflowing scattering clouds within sublimation radius are worth detailed follow-ups in the future.

astro-ph.GA↗

arXiv2019

Discovery of Metastable He I* $λ$10830 Mini-broad Absorption Lines and Very Narrow Paschen $α$ Emission Lines in the ULIRG Quasar IRAS F11119+3257

IRAS F11119+3257 is a quasar-dominated Ultra-Luminous InfraRed Galaxy, with a partially obscured narrow-line seyfert 1 nucleus. In this paper, we present the NIR spectroscopy of F11119+3257, in which we find unusual Paschen emission lines, and metastable He I* $λ$10830 absorption associated with the previously reported atomic sodium and molecular OH mini-BAL (Broad Absorption Line) outflow. Photo-ionization diagnosis confirms previous findings that the outflows are at kilo-parsec scales. Such large-scale outflows should produce emission lines. We indeed find that high-ionization emission lines ([O III], [Ne III], and [Ne V]) are dominated by blueshifted components at similar speeds to the mini-BALs. The blueshifted components are also detected in some low-ionization emission lines, such as [O II] $λ$3727 and some Balmer lines (H$α$, H$β$, and H$γ$), even though their cores are dominated by narrow ($FWHM_{\rm NEL} = 570\pm40$km s$^{-1}$) or broad components at the systemic redshift of $z=0.18966\pm0.00006$. The mass flow rate (230-730$~M_\odot \rm yr^{-1}$) and the kinetic luminosity ($\dot{E}_k \sim 10^{43.6-44.8} $erg s$^{-1}$) are then inferred jointly from the blueshifted emission and absorption lines. In the NIR spectrum of F11119+3257, we also find that the Paschen emission lines are unique, in which a very narrow ($FWHM=260\pm20~$km s$^{-1}$) component is shown in only Pa$α$. This narrow component most probably comes from heavily obscured star formation. Based on the Pa$α$ and Pa$β$ emissions, we obtain an extinction at the $H$ band, $A_H~>~2.1$ (or a reddenning of $E_{B-V}~>~$3.7), and a star formation rate of $SFR~>~130\rm M_\odot yr^{-1}$ that resembles the estimates inferred from the FIR emissions ($SFR_{\rm FIR} = 190\pm90$ $M_\odot$ yr$^{-1}$).

astro-ph.GA↗

arXiv2019

Fast Inflow Directly Feeding Black Hole Accretion Disk in Quasars

Quasars are high-luminosity active galactic nuclei believed to be powered by accretion of interstellar matter onto a super-massive black hole (SMBH) therein. Most of the observed energy is released in an accretion disk of inspiralling gas surrounding the SMBH. An enormous amount of fueling material is expected to be transported inwards. However, basic questions remain unanswered as to whether and how the accretion disks are supplied with external gas, since no disk-feeding inflow has hitherto been observed clearly. Here we report the discovery of highly redshifted broad absorption lines arising from neutral hydrogen and helium atoms in a small sample of quasars. Their absorption troughs show a broad range of Doppler velocities from zero extending continuously inward up to as high as $\sim 5,000$ km s$^{-1}$. We thus see through streams of cold gas moving with a radially inward velocity component that spans an immense gradient---a result of gravitational acceleration by the central SMBH. Extensive photo-ionization modeling for the archetypical object SDSS J103516.20+142200.6 indicates the inflowing gas to be dense, thick and moderately ionized, with a characteristic distance to the SMBH of $\sim 1,000$ gravitational radii, possibly overlapping or close to the outer accretion disk. Our results present the first compelling evidence for the long-sought inflow directly feeding quasars' accretion disks with external materials, likely originating from the dusty torus at a parsec scale. Our approach provides a new tool to probe the bulk of the so far elusive fueling inflows in quasars. Their studies may help address some of the fundamental questions concerning accretion physics, the onset and sustainment of quasar activity, and the SMBH growth at centers of most galaxies.

astro-ph.GA↗

arXiv2018

Low-mass Active Galactic Nuclei on the Fundamental Plane of Black Hole Activity

It is widely known that in active galactic nuclei (AGNs) and black hole X-ray binaries (BHXBs), there is a tight correlation among their radio luminosity ($L_R$), X-ray luminosity ($L_X$) and BH mass ($\mbh$), the so-called `fundamental plane' (FP) of BH activity. Yet the supporting data are very limited in the $\mbh$ regime between stellar mass (i.e., BHXBs) and 10$^{6.5}$\,\msun\ (namely, the lower bound of supermassive BHs in common AGNs). In this work, we developed a new method to measure the 1.4 GHz flux directly from the images of the VLA FIRST survey, and apply it to the type-1 low-mass AGNs in the \cite{2012ApJ...755..167D} sample. As a result, we obtained 19 new low-mass AGNs for FP research with both \mbh\ estimates ($\mbh \approx 10^{5.5-6.5}$\,\msun), reliable X-ray measurements, and (candidate) radio detections, tripling the number of such candidate sources in the literature.Most (if not all) of the low-mass AGNs follow the standard radio/X-ray correlation and the universal FP relation fitted with the combined dataset of BHXBs and supermassive AGNs by \citet{2009ApJ...706..404G}; the consistency in the radio/X-ray correlation slope among those accretion systems supports the picture that the accretion and ejection (jet) processes are quite similar in all accretion systems of different \mbh. In view of the FP relation, we speculate that the radio loudness $\mathcal{R}$ (i.e., the luminosity ratio of the jet to the accretion disk) of AGNs depends not only on Eddington ratio, but probably also on \mbh.

astro-ph.GA↗

arXiv2017

Reddening and He I* $\lambda10830$ Absorption Lines in Three Narrow-line Seyfert 1 Galaxies

We report the detection of heavy reddening and the He I* $\lambda10830$ absorption lines at the AGNs' redshift in three Narrow-line Seyfert 1 galaxies: SDSS J091848.61+211717.0, SDSS J111354.66+124439.0, and SDSS J122749.13+321458.9. They exhibit very red optical to near-infrared colors, narrow Balmer/Paschen broad emission lines and He I* $\lambda10830$ absorption lines. The ultraviolet-optical-infrared nucleus continua are reddened by the SMC extinction law of $E(B-V)\sim 0.74$, 1.17, and 1.24 mag for three objects, which are highly consistent with the values obtained from the broad-line Balmer decrements, but larger than those of narrow emission lines. The reddening analysis suggests the extinction dust simultaneously obscure the accretion disk, the broad emission line region, and the hot dust from the inner edge of the torus. It is possible that the dust obscuring the AGN structures are the dusty torus itself. Furthermore, the Cloudy analysis of the He I* $\lambda10830$ absorption lines propose the distance of the absorption materials to be the extend scale of the torus. That greatly increases probabilities of the obscure and absorption materials being the dusty torus.

astro-ph.GA↗

arXiv2017

Photoionization-driven Absorption Lines Variability in Balmer Absorption Line Quasar LBQS 1206+1052

In this paper we present an analysis of absorption line variability in mini-BAL quasar LBQS 1206+1052. The SDSS spectrum demonstrates that the absorption troughs can be divided into two components of blueshift velocities of $\sim$700 km s$^{-1}$ and $\sim$1400 km s$^{-1}$ relative to the quasar rest-frame. The former component shows rare Balmer absorption, which is an indicator of high density absorbing gas, thus the quasar is worth follow-up spectroscopic observations. Our follow-up optical and near-infrared spectra using MMT, YFOSC, TripleSpec and DBSP reveal that the strengths of the absorption lines vary for both of the two components, while the velocities do not change. We reproduce all of the spectral data by assuming that only the ionization state of the absorbing gas is variable and that all other physical properties are invariable. The variation of ionization is consistent with the variation of optical continuum from the V-band light-curve. Additionally, we can not interpret the data by assuming that the variability is due to a movement of the absorbing gas. Therefore, our analysis strongly indicates that the absorption line variability in LBQS 1206+1052 is photoionization-driven. As shown from photo-ionization simulations, the absorbing gas with blueshift velocity of $\sim$700 km s$^{-1}$ has a density in the range of $10^9$ to $10^{10}$ cm$^{-3}$ and a distance of $\sim$1 pc, and the gas with blueshift velocity of $\sim$1400 km s$^{-1}$ has a density of $10^3$ cm$^{-3}$ and a distance of $\sim$1 kpc.

astro-ph.GA↗

arXiv2017

Ultraviolet and Optical Emission-line Outflows in the Heavily Obscured Quasar SDSS J000610.67+121501.2: At the Scale of the Dusty Torus and Beyond

Broad emission-line outflows of active galactic nuclei (AGNs) have been proposed for many years but are very difficult to quantitatively study because of the coexistence of the gravitationally-bound and outflow emission. We present detailed analysis of a heavily reddened quasar, SDSS J000610.67+121501.2, whose normal ultraviolet (UV) broad emission lines (BELs) are heavily suppressed by the Dusty Torus as a natural "Coronagraph", thus the blueshifted BELs (BBELs) can be reliably measured. The physical properties of the emission-line outflows are derived as follows: ionization parameter $U \sim 10^{-0.5}$, column density $N_{\rm H}\sim 10^{22.0}$ cm$^{-2}$, covering fraction of $\sim 0.1$ and upper limit density of $n_{\rm H}\sim 10^{5.8}$ cm$^{-3}$. The outflow gases are located at least 41 pc away from the central engine, which suggests that they have expanded to the scale of the dust torus or beyond. Besides, Lya shows a narrow symmetric component, to our surprise, which is undetected in any other lines. After inspecting the narrow emission-line region and the starforming region as the origin of the Lya narrow line, we propose the end-result of outflows, diffusing gases in the larger region, acts as the screen of Lya photons. Future high spatial resolution spectrometry and/or spectropolarimetric observation are needed to make a final clarification.

astro-ph.GA↗

arXiv2015

Discovery of Extremely Broad Balmer Absorption Lines in SDSS J152350.42+391405.2

We present the discovery of Balmer line absorption from H$α$ to H$γ$ in an iron low-ionization broad absorption line (FeLoBAL) quasar SDSS J152350.42+391405.2 (hereafter J1523), by the quasi-simultaneous optical and near-infrared spectroscopy. The Balmer line absorption is at $z_{absor}$ = 0.6039 +/-0.0021 and blueshifted by v=10,353 km/s with respect to the Balmer emission lines. All Balmer BALs have uniform absorption profile with the widths of $Δ$ v ~12,000 km/s. We also found the absorption trough in He 1* $λ$10830 with the same velocity and width in the H-band TripleSpec spectrum of J1523. This object is only the tenth active galactic nucleus known to exhibit non-stellar Balmer absorption, and also the case with the highest velocity and broadest Balmer absorption lines which have ever been found. A CLOUDY analysis shows that the absorbers require an gas density of $log_{10} n_ e (cm^{-3})=9$ and an ionization parameter of $log_{10} U=-1.0$. They locate at a distance of ~0.2 pc from the central ionizing source which is slightly farther than that of BELRs. Furthermore, J1523 is one of the brightest Balmer-BAL quasar ever reported, with unique iron absorption variations, making it as the most promising candidate for follow up high-resolution spectroscopy, multi-band observations, and long-term monitoring.

astro-ph.GA↗

arXiv2015

Strong Variability of Overlapping Iron Broad Absorption Lines in Five Radio-selected Quasars

We present the variability study of broad absorption lines (BALs) in a uniformly radio-selected sample of 28 BAL quasars using the archival data from the FIRST Bright Quasar Survey (FBQS) and the Sloan Digital Sky Survey (SDSS), as well as those obtained by ourselves, covering time scales $\sim 1-10$ years in the quasar's rest-frame. To our surprise, 5 quasars showing strong variations are all belong to a special subclass of overlapping iron low ionization BAL (OFeLoBAL) quasars, however, other 4 non-overlapping FeLoBALs (non-OFeLoBALs) are invariable except one case with weak optical depth change. Meanwhile, we also find 6 typical variations of high-ionization and low-ionization BALs in this BAL quasar sample. Photoionization models suggest that OFeLoBALs are formed in a relative dense ($n_e>10^6$ cm$^{-3}$) outflows at a distance from the subparsec to the dozens of parsecs from the continuum source. They differ from those of non-OFeLoBALs, which are likely produced by low-density gas, locating at a distance of hundreds to thousands parsecs. Thus, OFeLoBALs and non-OFeLoBALs, i.e., FeLoBALs with/without strong BAL variations, perhaps represent the bimodality of Fe II absorption, the former is located in the active galactic nucleus environment rather than the host galaxy. We suggest that high density and small distance are the necessary conditions that cause OFeLoBALs. As suggested in the literature, strong BAL variability is possibly due to variability of the covering factor of BAL regions caused by clouds transiting across the line of sight rather than ionization variations.

astro-ph.GA↗

arXiv2014

Unshifted Metastable He I* Mini-Broad Absorption Line System in the Narrow Line Type 1 Quasar SDSS J080248.18$+$551328.9

We report the identification of an unusual absorption line system in the quasar SDSS J080248.18$+$551328.9 and present a detailed study of the system, incorporating follow-up optical and NIR spectroscopy. A few tens of absorption lines are detected, including He I*, Fe II* and Ni II* that arise from metastable or excited levels, as well as resonant lines in Mg I, Mg II, Fe II, Mn II, and Ca II. All of the isolated absorption lines show the same profile of width $Δv\sim 1,500$km s$^{-1}$ centered at a common redshift as that of the quasar emission lines, such as [O II], [S II], and hydrogen Paschen and Balmer series. With narrow Balmer lines, strong optical Fe II multiplets, and weak [O III] doublets, its emission line spectrum is typical for that of a narrow-line Seyfert 1 galaxy (NLS1). We have derived reliable measurements of the gas-phase column densities of the absorbing ions/levels. Photoionization modeling indicates that the absorber has a density of $n_{\rm H} \sim (1.0-2.5)\times 10^5~ {\rm cm}^{-3}$ and a column density of $N_{\rm H} \sim (1.0-3.2)\times 10^{21} \sim {\rm cm}^{-2}$, and is located at $R\sim100-250$ pc from the central super-massive black hole. The location of the absorber, the symmetric profile of the absorption lines, and the coincidence of the absorption and emission line centroid jointly suggest that the absorption gas is originated from the host galaxy and is plausibly accelerated by stellar processes, such as stellar winds \zhy{and/or} supernova explosions. The implications for the detection of such a peculiar absorption line system in an NLS1 are discussed in the context of co-evolution between super-massive black hole growth and host galaxy build-up.

astro-ph.GA↗

arXiv2013

Anomalously Steep Reddening Law in Quasars: An Exceptional Example Observed in IRAS14026+4341

A fraction of the heavily reddened quasars require a reddening curve which is even steeper than that of the Small Magellanic Cloud. In this paper, we thoroughly characterize the anomalously steep reddening law in quasars, via an exceptional example observed in IRAS 14026+4341. By comparing the observed spectrum to the quasar composite spectrum, we derive a reddening curve in the rest-frame wavelength range of 1200 Å--10000 Å. It is featured with a steep rise at wavelengths shorter than 3000 Å, but no significant reddening at longer wavelengths. The absence of dust reddening in optical continuum is confirmed by the normal broad-line Balmer decrement (the H$α$/H$β$ ratio) in IRAS 14026+4341. The anomalous reddening curve can be satisfactorily reproduced by a dust model containing silicate grains in a power-law size distribution, $dn(a)/da \propto a^{-1.4}$, truncated at a maximum size $a_{max}=70 {\rm nm}$. The unusual size distribution may be caused by the destruction of large "stardust" grains by quasar activities or a different dust formation mechanism (i.e., the in situ formation of dust grains in quasar outflows). It is also possible that the analogies of the dust grains observed toward the Galactic center is responsible for the steep reddening curve. In addition, we find that IRAS 14026+4341 is a weak emission-line quasar (i.e., PHL 1811 analogies) with heavy dust reddening and blueshifted broad absorption lines.

astro-ph.CO↗