A scalable algorithm for structure identification of complex gene regulatory network from temporal expression data.

BACKGROUND: Gene regulatory interactions are of fundamental importance to various biological functions and processes. However, only a few previous computational studies have claimed success in revealing genome-wide regulatory landscapes from temporal gene expression data, especially for complex eukaryotes like human. Moreover, recent work suggests that these methods still suffer from the curse of dimensionality if a network size increases to 100 or higher. RESULTS: Here we present a novel scalable algorithm for identifying genome-wide gene regulatory network (GRN) structures, and we have verified the algorithm performances by extensive simulation studies based on the DREAM challenge benchmark data. The highlight of our method is that its superior performance does not degenerate even for a network size on the order of 104, and is thus readily applicable to large-scale complex networks. Such a breakthrough is achieved by considering both prior biological knowledge and multiple topological properties (i.e., sparsity and hub gene structure) of complex networks in the regularized formulation. We also validate and illustrate the application of our algorithm in practice using the time-course gene expression data from a study on human respiratory epithelial cells in response to influenza A virus (IAV) infection, as well as the CHIP-seq data from ENCODE on transcription factor (TF) and target gene interactions. An interesting finding, owing to the proposed algorithm, is that the biggest hub structures (e.g., top ten) in the GRN all center at some transcription factors in the context of epithelial cell infection by IAV. CONCLUSIONS: The proposed algorithm is the first scalable method for large complex network structure identification. The GRN structure identified by our algorithm could reveal possible biological links and help researchers to choose which gene functions to investigate in a biological event. The algorithm described in this article is implemented in MATLAB Ⓡ , and the source code is freely available from https://github.com/Hongyu-Miao/DMI.git .

Segmentation of pelvic structures in T2 MRI via MR-to-CT synthesis.

Segmentation of multiple pelvic structures in MRI volumes is a prerequisite for many clinical applications, such as sarcopenia assessment, bone density measurement, and muscle-to-fat volume ratio estimation. While many CT-specific datasets and automated CT-based multi-structure pelvis segmentation methods exist, there are few MRI-specific multi-structure segmentation methods in literature. In this pilot work, we propose a lightweight and annotation-free pipeline to synthetically translate T2 MRI volumes of the pelvis to CT, and subsequently leverage an existing CT-only tool called TotalSegmentator to segment 8 pelvic structures in the generated CT volumes. The predicted masks were then mapped back to the original MR volumes as segmentation masks. We compared the predicted masks against the expert annotations of the public TCGA-UCEC dataset and an internal dataset. Experiments demonstrated that the proposed pipeline achieved Dice measures ≥65% for 8 pelvic structures in T2 MRI. The proposed pipeline is an alternative method to obtain multi-organ and structure segmentations without being encumbered by time-consuming manual annotations. By exploiting the significant research progress in CTs, it is possible to extend the proposed pipeline to other MRI sequences in principle. Our research bridges the chasm between the current CT-based multi-structure segmentation and MRI-based segmentation. The manually segmented structures in the TCGA-UCEC dataset are publicly available.

Dynamic analysis of sheep Brucellosis model with environmental infection pathways.

We develop a mathematical model for the transmission of brucellosis in sheep taking into account external inputs, immunity, stage structure and other factors. We find the the basic reproduction number $ R_0 $ in terms of the model parameters, and prove the global stability of the disease-free equilibrium. Then, the existence and global stability of the endemic equilibrium is proven. Finally, sheep data from Yulin, China are employed to fit the model parameters for three different environmental infection exposure conditions. The variability between different models in terms of control measures are analyzed numerically. Results show that the model is sensitive to the control parameters for different environmental infection exposure functions. This means that in practical modeling, the selection of environmental infection exposure functions needs to be properly considered.

Relation between adhesiveness and surface leachate rheological properties of cooked noodles: From the view of starch fine molecular structure.

Adhesiveness is one of the important sensory properties of noodles. A better understanding of the molecular basis of the sensory property would allow for a more reasonable selection of wheat ingredients. In this study, the adhesiveness of cooked noodles made from reconstituted wheat flour was evaluated using a texture profile analyzer. The rheological properties of dissolved solids from the noodle surface were measured. The composition and structure of leachate, as well as multi-structure of native starch were analyzed. The correlation analysis showed that adhesiveness positively correlates with the loss modulus parameters [log K' (consistency coefficient) and Ea (activation energy)]. Increased content of leached starch can ascend the rheological parameters (log K' and Ea). In addition, more short amylose chains with 100-500 degree of polymerization (DP) in leached starch also induced the increase of log K' (p < 0.01) and Ea (p < 0.01). The fraction of short amylose chains in leached starch was negatively correlated with the content of long amylopectin chains ranged from 37 to 100 DP in native starch (p < 0.05), as well as the median diameter of particles (p < 0.01). Multi-level structures of native starch granules largely determine the composition and molecular structure of noodle leachate, consequently contributing to surface rheological properties. It is thus proposed that wheat starch with more long amylopectin chains and large granules could be used to reduce noodle adhesiveness. Moreover, the rheological approach using noodle surface materials proposed here is useful to predict the noodle adhesiveness as an important sensory characteristic.

Data Size and Quality Matter: Generating Physically-Realistic Distance Maps of Protein Tertiary Structures.

With the debut of AlphaFold2, we now can get a highly-accurate view of a reasonable equilibrium tertiary structure of a protein molecule. Yet, a single-structure view is insufficient and does not account for the high structural plasticity of protein molecules. Obtaining a multi-structure view of a protein molecule continues to be an outstanding challenge in computational structural biology. In tandem with methods formulated under the umbrella of stochastic optimization, we are now seeing rapid advances in the capabilities of methods based on deep learning. In recent work, we advance the capability of these models to learn from experimentally-available tertiary structures of protein molecules of varying lengths. In this work, we elucidate the important role of the composition of the training dataset on the neural network's ability to learn key local and distal patterns in tertiary structures. To make such patterns visible to the network, we utilize a contact map-based representation of protein tertiary structure. We show interesting relationships between data size, quality, and composition on the ability of latent variable models to learn key patterns of tertiary structure. In addition, we present a disentangled latent variable model which improves upon the state-of-the-art variable autoencoder-based model in key, physically-realistic structural patterns. We believe this work opens up further avenues of research on deep learning-based models for computing multi-structure views of protein molecules.

Regularized multi-structural shape modeling of the knee complex based on deep functional maps.

The incorporation of a-priori knowledge on the shape of anatomical structures and their variation through Statistical Shape Models (SSMs) has shown to be very effective in guiding highly uncertain image segmentation problems. In this paper, we construct multiple-structure SSMs of purely geometric nature, that describe the relationship between adjacent anatomical components through Canonical Correlation Analysis. Shape inference is then conducted based on a regularization term on the shape likelihood providing more reliable structure representations. A fundamental prerequisite for performing statistical shape analysis on a set of objects is the identification of corresponding points on their associated surfaces. We address the correspondence problem using the recently proposed Functional Maps framework, which is a generalization of point-to-point correspondence to manifolds. Additionally, we show that, by incorporating techniques from the deep learning theory into this framework, we can further enhance the ability of SSMs to better capture the shape variation in a given dataset. The efficiency of our approach is illustrated through the creation of 3D models of the human knee complex in two application scenarios: incomplete or noisy shape reconstruction and missing structure estimation.

The multi-scale structure, thermal and digestion properties of mung bean starch.

Identification and selection one special variety mung bean for lower GI food is very useful, however, the fundamental study for mung bean starch is still insufficient to meet its demand. In this study, four varieties of mostly planted mung bean in China were selected as model materials. The multi-scale structure of mung bean starch was characterized by SEC, HPAEC, XRD, SAXS, and SEM. SEC and HPAEC give the amylose contents, amylose and amylopectin fine structure of mung bean starch. Mung bean starch from XRD spectrum display CA type semi crystallinity. The crystalline lamellar thickness from SAXS curves were 7.34-7.60 nm. DSC indicated that the peak gelatinization temperature is at 67 °C-68 °C. Resistant starch in mung bean disappears rapidly after cooking, although the amount of slowly digested starch was still more than half of the total starch. Since the gene backgrounds of the mung bean starch samples are very close, there was no obvious difference in their molecular and aggregated state structure, and the digestion properties were similar, too. Unique SEC and HPAEC profiles of starch chain length distribution can be utilized to help find more genetic resources and cultivate variety to meet the needs for starch applications.

Understanding the digestibility of rice starch-gallic acid complexes formed by high pressure homogenization.

Rice starch-Gallic acid (GA) complex (RSP-GA) was prepared by high pressure homogenization (HPH), and the effect of GA on the digestibility and multi-structure of rice starch under HPH was investigated. The results showed that, after HPH, the digestibility of starch substantially changed in the reduced rapidly digestible starch (RDS) content, and increased resistant starch (RS) after interacting with GA. In particular, the RS content of RSP-GA ranged from 5.4% to 29.7%, which were much higher than that of rice starch (1.6%). Meanwhile, the results indicated that rice starch and GA were aggregated by hydrogen bonding and van der Waals forces to form a single helix V7 type complex during HPH processing. Moreover, with the increase addition of GA, the fractal structure of the RSP-GA is converted into a mass fractal structure, and the aggregate structure gradually became compact due to the enhancement of rearrangement and aggregation behavior of the degraded starch molecular chains. It thus reduced the accessibility of the starch molecules to digestive enzymes. These results demonstrated that HPH and GA complexation could be beneficial to control the digestion of starch products with desired digestibility.

Multi-structure docking analysis of BACE1 crystal structures and non-peptidic ligands.

In order to design novel non-peptidic inhibitors of BACE1, many research groups have attempted using computational studies including docking analyses. Since there are too many 3D structures for BACE1 in the protein database, the selection of suitable crystal structures is a key prerequisite for the successful application of molecular docking. We employed a multi-structure docking protocol. In which 615 ligands' structures were docked into 150 BACE1 structures. The large number of the resultant docking scores were post-processed by different data analysis methods including exploratory data analysis, regression analysis and discriminant analysis. It was found that using one crystal structure for docking did not result in high accuracy for predicting activity of the BACE1 inhibitors. Instead, using of the multi-structural docking scores, post-processed by chemometrics methods arrived to highly accurate predictive models. In this regards, the PDB accession codes of 4B70, 4DVF and 2WEZ could discriminate between active and inactive compounds, with higher accuracy. Clustering of the BACE1 structures based on principal component analysis of the crystallographic structures the revealed that the discriminant structures are in the center of the clusters. Thus, these structures can be selected as predominant crystal structures for docking studies of non-peptidic BACE1 inhibitors.

Computational design of zinc-ion-responsive two-photon fluorescent probes with conjugated multi-structures.

A series of conjugated multi-structured fluorescent probe molecules based on a salen ligand were designed and investigated in dimethyl sulfoxide solvent using a quantum-chemical method. The results indicate that the one-photon absorption and fluorescence emission spectra (λ (O) and λ (EM)) of these molecules generally show redshifts (of 23.1-74.5 and 22.7-116.6 nm, respectively) upon the coordination of the molecules to Zn(2+). Large Stokes shifts (1511.2-11744.1 cm(-1)) were found for the molecules, meaning that interference between λ (O) and λ (EM) can be avoided for these molecules. The two-photon absorption spectra of the molecules usually present blueshifts, but the two-photon absorption cross-section (δ) greatly increases (by 221.5-868.0 GM) upon the coordination of the molecules with Zn(2+). Most of the molecules show strong two-photon absorption peaks in the range 678.2-824.4 nm, i.e., in the near-infrared region. In a word, the expanded π-conjugated frameworks of these molecules lead to redshifted λ (O) and λ (EM) and enhanced δ values. Moreover, (L-phenyl)​2 and (L-phenyl-ethynyl)2 are the most suitable of the multi-structured molecules examined in this work for use as two-photon fluorescent probes for zinc ion detection in vivo. Graphical Abstract Scheme of the calculated transition energies (E0k and E0n) and the transition dipole moments (M0k and Mkn). NTO 109, NTO 197 and NTO 228 of Zn(L-phenyl-ethynyl), Zn2(L-phenyl-ethynyl)2 and Zn3(L-phenyl)3 for one-photon absorption, respectively.