MCPNet: a parallel maximum capacity-based genome-scale gene network construction framework.

MOTIVATION: Gene network reconstruction from gene expression profiles is a compute- and data-intensive problem. Numerous methods based on diverse approaches including mutual information, random forests, Bayesian networks, correlation measures, as well as their transforms and filters such as data processing inequality, have been proposed. However, an effective gene network reconstruction method that performs well in all three aspects of computational efficiency, data size scalability, and output quality remains elusive. Simple techniques such as Pearson correlation are fast to compute but ignore indirect interactions, while more robust methods such as Bayesian networks are prohibitively time consuming to apply to tens of thousands of genes. RESULTS: We developed maximum capacity path (MCP) score, a novel maximum-capacity-path-based metric to quantify the relative strengths of direct and indirect gene-gene interactions. We further present MCPNet, an efficient, parallelized gene network reconstruction software based on MCP score, to reverse engineer networks in unsupervised and ensemble manners. Using synthetic and real Saccharomyces cervisiae datasets as well as real Arabidopsis thaliana datasets, we demonstrate that MCPNet produces better quality networks as measured by AUPRC, is significantly faster than all other gene network reconstruction software, and also scales well to tens of thousands of genes and hundreds of CPU cores. Thus, MCPNet represents a new gene network reconstruction tool that simultaneously achieves quality, performance, and scalability requirements. AVAILABILITY AND IMPLEMENTATION: Source code freely available for download at https://doi.org/10.5281/zenodo.6499747 and https://github.com/AluruLab/MCPNet, implemented in C++ and supported on Linux.

EnGRaiN: a supervised ensemble learning method for recovery of large-scale gene regulatory networks.

MOTIVATION: Reconstruction of genome-scale networks from gene expression data is an actively studied problem. A wide range of methods that differ between the types of interactions they uncover with varying trade-offs between sensitivity and specificity have been proposed. To leverage benefits of multiple such methods, ensemble network methods that combine predictions from resulting networks have been developed, promising results better than or as good as the individual networks. Perhaps owing to the difficulty in obtaining accurate training examples, these ensemble methods hitherto are unsupervised. RESULTS: In this article, we introduce EnGRaiN, the first supervised ensemble learning method to construct gene networks. The supervision for training is provided by small training datasets of true edge connections (positives) and edges known to be absent (negatives) among gene pairs. We demonstrate the effectiveness of EnGRaiN using simulated datasets as well as a curated collection of Arabidopsis thaliana datasets we created from microarray datasets available from public repositories. EnGRaiN shows better results not only in terms of receiver operating characteristic and PR characteristics for both real and simulated datasets compared with unsupervised methods for ensemble network construction, but also generates networks that can be mined for elucidating complex biological interactions. AVAILABILITY AND IMPLEMENTATION: EnGRaiN software and the datasets used in the study are publicly available at the github repository: https://github.com/AluruLab/EnGRaiN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Reverse engineering and analysis of large genome-scale gene networks.

Reverse engineering the whole-genome networks of complex multicellular organisms continues to remain a challenge. While simpler models easily scale to large number of genes and gene expression datasets, more accurate models are compute intensive limiting their scale of applicability. To enable fast and accurate reconstruction of large networks, we developed Tool for Inferring Network of Genes (TINGe), a parallel mutual information (MI)-based program. The novel features of our approach include: (i) B-spline-based formulation for linear-time computation of MI, (ii) a novel algorithm for direct permutation testing and (iii) development of parallel algorithms to reduce run-time and facilitate construction of large networks. We assess the quality of our method by comparison with ARACNe (Algorithm for the Reconstruction of Accurate Cellular Networks) and GeneNet and demonstrate its unique capability by reverse engineering the whole-genome network of Arabidopsis thaliana from 3137 Affymetrix ATH1 GeneChips in just 9 min on a 1024-core cluster. We further report on the development of a new software Gene Network Analyzer (GeNA) for extracting context-specific subnetworks from a given set of seed genes. Using TINGe and GeNA, we performed analysis of 241 Arabidopsis AraCyc 8.0 pathways, and the results are made available through the web.

A brassinosteroid transcriptional network revealed by genome-wide identification of BESI target genes in Arabidopsis thaliana.

Brassinosteroids (BRs) are important regulators for plant growth and development. BRs signal to control the activities of the BES1 and BZR1 family transcription factors. The transcriptional network through which BES1 and BZR regulate large number of target genes is mostly unknown. By combining chromatin immunoprecipitation coupled with Arabidopsis tiling arrays (ChIP-chip) and gene expression studies, we have identified 1609 putative BES1 target genes, 404 of which are regulated by BRs and/or in gain-of-function bes1-D mutant. BES1 targets contribute to BR responses and interactions with other hormonal or light signaling pathways. Computational modeling of gene expression data using Algorithm for the Reconstruction of Accurate Cellular Networks (ARACNe) reveals that BES1-targeted transcriptional factors form a gene regulatory network (GRN). Mutants of many genes in the network displayed defects in BR responses. Moreover, we found that BES1 functions to inhibit chloroplast development by repressing the expression of GLK1 and GLK2 transcription factors, confirming a hypothesis generated from the GRN. Our results thus provide a global view of BR regulated gene expression and a GRN that guides future studies in understanding BR-regulated plant growth.

Electrospun nanofibrous membranes for temperature regulation of microfluidic seed growth chips.

This paper reports a simple and effective approach to control statistical pore size distributions within electrospun nanofibrous membranes (ENMs), by choosing appropriate spinning times of electro-spinning deposition. Mean pore diameter of ENMs decreases exponentially with increasing spinning time. This pore-size control method is demonstrated to regulate amount of heat energy reaching microfluidic seed growth chips (SGC) and thus growth temperature of seeds on the chips, without using sophisticated semiconductor manufacturing techniques or additional on-chip electronic circuits. Decreasing mean pore diameter of ENMs causes to decrease the on-chip temperature, following a second-order polynomial trend. Phenotypic study based on real-time observation of root architecture is conducted on multiple SGCs under various temperature conditions obtained by using ENMs with different pore size distributions.

Conserved active site sequences in Arabidopsis plastid terminal oxidase (PTOX): in vitro and in planta mutagenesis studies.

The plastid terminal oxidase (PTOX) is distantly related to the mitochondrial alternative oxidase (AOX). Both are members of the diiron carboxylate quinol oxidase (DOX) class of proteins. PTOX and AOX contain 20 highly conserved amino acids, six of which are Fe-binding ligands. We have previously used in vitro and in planta activity assays to examine the functional importance of the Fe-binding sites. In this report, we conduct alanine-scanning mutagenesis on the 14 other conserved sites using our in vitro and in planta assay procedures. We found that the 14 sites fall into three classes: (i) Ala-139, Pro-142, Glu-171, Asn-174, Leu-179, Pro-216, Ala-230, Asp-287, and Arg-293 are dispensable for activity; (ii) Tyr-234 and Asp-295 are essential for activity; and (iii) Leu-135, His-151, and Tyr-212 are important but not essential for activity. Our data are consistent with the proposed role of some of these residues in active site conformation, substrate binding, and/or catalysis. Titration experiments showed that down-regulation of PTOX to approximately 3% of wild-type levels did not compromise plant growth, at least under ambient growth conditions. This suggests that PTOX is normally in excess, especially early in thylakoid membrane biogenesis.

Generation of transgenic maize with enhanced provitamin A content.

Vitamin A deficiency (VAD) affects over 250 million people worldwide and is one of the most prevalent nutritional deficiencies in developing countries, resulting in significant socio-economic losses. Provitamin A carotenoids such as beta-carotene, are derived from plant foods and are a major source of vitamin A for the majority of the world's population. Several years of intense research has resulted in the production of 'Golden Rice 2' which contains sufficiently high levels of provitamin A carotenoids to combat VAD. In this report, the focus is on the generation of transgenic maize with enhanced provitamin A content in their kernels. Overexpression of the bacterial genes crtB (for phytoene synthase) and crtI (for the four desaturation steps of the carotenoid pathway catalysed by phytoene desaturase and zeta-carotene desaturase in plants), under the control of a 'super gamma-zein promoter' for endosperm-specific expression, resulted in an increase of total carotenoids of up to 34-fold with a preferential accumulation of beta-carotene in the maize endosperm. The levels attained approach those estimated to have a significant impact on the nutritional status of target populations in developing countries. The high beta-carotene trait was found to be reproducible over at least four generations. Gene expression analyses suggest that increased accumulation of beta-carotene is due to an up-regulation of the endogenous lycopene beta-cylase. These experiments set the stage for the design of transgenic approaches to generate provitamin A-rich maize that will help alleviate VAD.

Mitogen-activated protein kinases, Fus3 and Kss1, regulate chronological lifespan in yeast.

Using a systems-based approach, we have identified several genes not previously evaluated for a role(s) in chronological aging. Here, we have thoroughly investigated the chronological lifespan (CLS) of three of these genes (FUS3, KSS1 and HOG1) and their protein products, each of which have well-defined cell signaling roles in young cells. The importance of FUS3 and KSS1 in CLS are largely unknown and analyzed here for the first time. Using both qualitative and quantitative CLS assays, we show that deletion of any of the three MAPK's increases yeast lifespan. Furthermore, combined deletion of any MAPK and TOR1, most prominently fus3Δ/tor1Δ, produces a two-stage CLS response ending in lifespan increase greater than that of tor1Δ. Similar effects are achieved upon endogenous expression of a non-activatable form of Fus3. We speculate that the autophagy-promoting role of FUS3, which is inherently antagonistic to the role of TOR1, may in part be responsible for the differential aging phenotype of fus3Δ/tor1Δ. Consistent with this notion we show that nitrogen starvation, which promotes autophagy by deactivating Tor1, results in decreased CLS if FUS3 is deleted. Taken together, these results reveal a previously unrealized effect of mating-specific MAPKs in the chronological lifespan of yeast.

RD26 mediates crosstalk between drought and brassinosteroid signalling pathways.

Brassinosteroids (BRs) regulate plant growth and stress responses via the BES1/BZR1 family of transcription factors, which regulate the expression of thousands of downstream genes. BRs are involved in the response to drought, however the mechanistic understanding of interactions between BR signalling and drought response remains to be established. Here we show that transcription factor RD26 mediates crosstalk between drought and BR signalling. When overexpressed, BES1 target gene RD26 can inhibit BR-regulated growth. Global gene expression studies suggest that RD26 can act antagonistically to BR to regulate the expression of a subset of BES1-regulated genes, thereby inhibiting BR function. We show that RD26 can interact with BES1 protein and antagonize BES1 transcriptional activity on BR-regulated genes and that BR signalling can also repress expression of RD26 and its homologues and inhibit drought responses. Our results thus reveal a mechanism coordinating plant growth and drought tolerance.

Microarray Data Processing Techniques for Genome-Scale Network Inference from Large Public Repositories.

Pre-processing of microarray data is a well-studied problem. Furthermore, all popular platforms come with their own recommended best practices for differential analysis of genes. However, for genome-scale network inference using microarray data collected from large public repositories, these methods filter out a considerable number of genes. This is primarily due to the effects of aggregating a diverse array of experiments with different technical and biological scenarios. Here we introduce a pre-processing pipeline suitable for inferring genome-scale gene networks from large microarray datasets. We show that partitioning of the available microarray datasets according to biological relevance into tissue- and process-specific categories significantly extends the limits of downstream network construction. We demonstrate the effectiveness of our pre-processing pipeline by inferring genome-scale networks for the model plant Arabidopsis thaliana using two different construction methods and a collection of 11,760 Affymetrix ATH1 microarray chips. Our pre-processing pipeline and the datasets used in this paper are made available at http://alurulab.cc.gatech.edu/microarray-pp.

Control of chloroplast redox by the IMMUTANS terminal oxidase.

Variegation mutants offer excellent opportunities to study interactions between the nucleus-cytoplasm, the chloroplast, and the mitochondrion. Variegation in the immutans (im) mutant of Arabidopsis is induced by a nuclear recessive gene and the extent of variegation can be modulated by light and temperature. Whereas the green sectors have morphologically normal chloroplasts, the white sectors are devoid of pigments and accumulate a colourless carotenoid, phytoene. The green sectors are hypothesized to arise from cells that have avoided irreversible photooxidative damage whereas the white sectors originate from cells that are photooxidized. Cloning of the IMMUTANS (IM) gene has revealed that IMMUTANS (IM) is a plastid homologue of the mitochondrial alternative oxidase. This finding suggested a model in which IM functions as a redox component of the phytoene desaturation pathway, which requires phytoene desaturase activity. Consistent with this idea, IM has quinol oxidase activity in vitro. Recent studies have revealed that IM plays a more global role in plastid metabolism. For example, it appears to be the elusive terminal oxidase of chlororespiration and also functions as a light stress protein.

Plant chip for high-throughput phenotyping of Arabidopsis.

We report on the development of a vertical and transparent microfluidic chip for high-throughput phenotyping of Arabidopsis thaliana plants. Multiple Arabidopsis seeds can be germinated and grown hydroponically over more than two weeks in the chip, thus enabling large-scale and quantitative monitoring of plant phenotypes. The novel vertical arrangement of this microfluidic device not only allows for normal gravitropic growth of the plants but also, more importantly, makes it convenient to continuously monitor phenotypic changes in plants at the whole organismal level, including seed germination and root and shoot growth (hypocotyls, cotyledons, and leaves), as well as at the cellular level. We also developed a hydrodynamic trapping method to automatically place single seeds into seed holding sites of the device and to avoid potential damage to seeds that might occur during manual loading. We demonstrated general utility of this microfluidic device by showing clear visible phenotypes of the immutans mutant of Arabidopsis, and we also showed changes occurring during plant-pathogen interactions at different developmental stages. Arabidopsis plants grown in the device maintained normal morphological and physiological behaviour, and distinct phenotypic variations consistent with a priori data were observed via high-resolution images taken in real time. Moreover, the timeline for different developmental stages for plants grown in this device was highly comparable to growth using a conventional agar plate method. This prototype plant chip technology is expected to lead to the establishment of a powerful experimental and cost-effective framework for high-throughput and precise plant phenotyping.

Mechanisms and networks for brassinosteroid regulated gene expression.

Brassinosteroids (BRs) signal through plasma membrane-localized receptor BRI1 and other components including negatively acting BIN2 kinase to regulate BES1/BZR1 family transcription factors, which controls the expression of thousands of genes for various BR responses. Recent studies demonstrated that BR regulation of gene expression involves histone-modifying enzymes and changes of chromatin structure. Genomic experiments identified a few thousand BES1/BZR1 target genes, many of which are involved in plant growth and various signaling pathways. Moreover, BES1/BZR1 interact with many transcription factors to integrate BR and other signaling pathways. Finally, in addition to regulating BES1/BZR1, BIN2 can phosphorylate and regulate the activities of more transcription factors and signaling components, providing additional inputs of BR signaling to the BR transcriptional network and points of crosstalk with different pathways.

Genetic modification of low phytic acid 1-1 maize to enhance iron content and bioavailability.

High phytate content in staple food crops is a major barrier to successful iron biofortification. We have exploited the low phytic acid 1-1 (lpa1-1) mutant of maize to generate transgenic plants with up-to 70 µg/g seed iron through the endosperm-specific overexpression of soybean ferritin, resulting in more than 2-fold improvement in iron bioavailability. The levels of bioavailable seed iron achieved in this study greatly exceed any achieved thus far and closely approach values estimated to have a nutritional impact on target populations. Gene expression studies reveal a large induction of the YS1 transporter in leaves and severe repression of an iron acquisition gene DMAS1 in roots, suggesting significant alterations in the iron homeostatic mechanisms in transgenic lpa1-1. Furthermore, preliminary tests show that the high-iron lpa1-1 seeds have higher germination rates and seedling vigor when compared to those of the nontransgenic seeds, which may help improve their value to plant breeders.

PDS activity acts as a rheostat of retrograde signaling during early chloroplast biogenesis.

Chloroplasts are crucial for the process of photosynthesis, as well as for developmental and environmental sensing. One of the important mechanisms of sensing is retrograde (plastid-to-nucleus) signaling, whereby the state of the chloroplast is signaled to the nucleus, resulting in alterations in gene expression for chloroplast proteins, usually at the transcriptional level. Retrograde signaling was early studied in carotenoid-deficient plants that contain, upon exposure to high light, photooxidized plastids that arise because of an inability to quench ROS produced during the light reactions of photosynthesis. Phytoene desaturase (PDS) is required for one of the early steps of the carotenogenic pathway, and impaired PDS activity during early chloroplast biogenesis results in a highly reduced plastoquinone pool (high excitation pressure), accumulation of the colorless C(40) intermediate, phytoene, and white photooxidized plastids. Here, we discuss results from global transcript profiling of white leaf tissues of Arabidopsis that are blocked at the PDS step in three different ways--two by mutation (immutans & pds3) and one by inhibitor treatment (norflurazon). We show that the molecular phenotypes of the three tissues bear many similarities, but that there are also significant tissue-specific differences. We propose that PDS acts as a rheostat of excitation pressure-mediated retrograde signaling during chloroplast development, and speculate that whether the rheostat is set high (as in pds3 and NF-treated seedlings), intermediate (as in im) or low (as in WT) is a crucial determinant of the suite of genes that is expressed during chloroplast biogenesis.

Quantification of ferritin from staple food crops.

Ferritin-iron has been shown to be as bioavailable as ferrous sulfate in humans. Thus, biofortification to breed crops with high ferritin content is a promising strategy to alleviate the global iron deficiency problem. Although ferritin is present in all food crops, its concentration varies between species and varieties. Therefore, a successful ferritin biofortification strategy requires a method to rapidly measure ferritin concentrations in food crops. The objective of this study was to develop a simple and reliable ELISA using an anti-ferritin polyclonal antibody to detect ferritin in various crops. Crude seed extracts were found to have 10.2 +/- 1.0, 4.38 +/- 0.9, 1.2 +/- 0.3, 0.38 +/- 0.1, and 0.04 +/- 0.01 microg of ferritin/g of dry seed in red beans, white beans, wheat, maize, and brown rice, respectively. Although the measured absolute concentrations of ferritin values were low, the presented method is applicable for rapid screening for the relative ferritin concentrations of large numbers of seeds to identify and breed ferritin-rich crops.

Chloroplast photooxidation-induced transcriptome reprogramming in Arabidopsis immutans white leaf sectors.

Arabidopsis (Arabidopsis thaliana) immutans (im) has green and white sectoring due to the action of a nuclear recessive gene, IMMUTANS. The green sectors contain normal-appearing chloroplasts, whereas the white sectors contain abnormal chloroplasts that lack colored carotenoids due to a defect in phytoene desaturase activity. Previous biochemical and molecular characterizations of the green leaf sectors revealed alterations suggestive of a source-sink relationship between the green and white sectors of im. In this study, we use an Affymetrix ATH1 oligoarray to further explore the nature of sink metabolism in im white tissues. We show that lack of colored carotenoids in the im white tissues elicits a differential response from a large number of genes involved in various cellular processes and stress responses. Gene expression patterns correlate with the repression of photosynthesis and photosynthesis-related processes in im white tissues, with an induction of Suc catabolism and transport, and with mitochondrial electron transport and fermentation. These results suggest that energy is derived via aerobic and anaerobic metabolism of imported sugar in im white tissues for growth and development. We also show that oxidative stress responses are largely induced in im white tissues; however, im green sectors develop additional energy-dissipating mechanisms that perhaps allow for the formation of green sectors. Furthermore, a comparison of the transcriptomes of im white and norflurazon-treated white leaf tissues reveals global as well as tissue-specific responses to photooxidation. We conclude that the differences in the mechanism of phytoene desaturase inhibition play an important role in differentiating these two white tissues.

Alterations in photosynthesis in Arabidopsis lacking IMMUTANS, a chloroplast terminal oxidase.

Green and white variegation in the Arabidopsis immutans (im) mutant is caused by a nuclear recessive gene. The green sectors contain cells with normal-appearing chloroplasts, while cells in the white sectors have photooxidized plastids lacking organized lamellae. In the present experiments, we found that the green im sectors have enhanced rates of carbon assimilation (monitored by (14)CO(2) uptake) and that there are corresponding increases in the activities of Rubisco and SPS, elevated starch and sucrose pool sizes, and an altered pattern of carbohydrate partitioning that favors sucrose over starch. We hypothesize that these increases are due, at least in part, to interactions with white sectors, perhaps to compensate for reductions in total source tissue. Consistent with this idea, the im white sectors accumulate low levels of sucrose and acid invertase activities are markedly increased in the white versus green cells. This suggests that there is a sucrose gradient between the green and white sectors, and that sucrose is transported from the green to white cells in response to sink demand. The expression of photosynthetic genes is not appreciably altered in the green im sectors versus wild type, but rather there is an up-regulation of genes involved in defense against oxidative stress and down-regulation of genes involved in cell wall biosynthesis. We postulate that changes in photosynthesis in the im green cells are driven by a need for photoprotection (especially early in chloroplast biogenesis) and due to source-sink interactions.

Variegation mutants and mechanisms of chloroplast biogenesis.

Variegated plants typically have green- and white-sectored leaves. Cells in the green sectors contain normal-appearing chloroplasts, whereas cells in the white sectors lack pigments and appear to be blocked at various stages of chloroplast biogenesis. Variegations can be caused by mutations in nuclear, chloroplast or mitochondrial genes. In some plants, the green and white sectors have different genotypes, but in others they have the same (mutant) genotype. One advantage of variegations is that they provide a means of studying genes for proteins that are important for chloroplast development, but for which mutant analysis is difficult, either because mutations in a gene of interest are lethal or because they do not show a readily distinguishable phenotype. This paper focuses on Arabidopsis variegations, for which the most information is available at the molecular level. Perhaps the most interesting of these are variegations caused by defective nuclear gene products in which the cells of the mutant have a uniform genotype. Two questions are of paramount interest: (1) What is the gene product and how does it function in chloroplast biogenesis? (2) What is the mechanism of variegation and why do green sectors arise in plants with a uniform (mutant) genotype? Two paradigms of variegation mechanism are described: immutans (im) and variegated2 (var2). Both mechanisms emphasize compensating activities and the notion of plastid autonomy, but redundant gene products are proposed to play a role in var2, but not in im. It is hypothesized that threshold levels of certain activities are necessary for normal chloroplast development.

Arabidopsis variegation mutants: new insights into chloroplast biogenesis.

Plant variegations are characterized by the presence of white sectors in normally green tissues and organs. Whereas the white sectors contain defective plastids that lack coloured pigments, the green sectors contain morphologically normal chloroplasts. Variegation mutants are defective in chloroplast developmental processes and arise due to mutations in nuclear or organellar genes. Despite their widespread occurrence in nature, only a few variegations have been studied at the molecular level. In this review, recent progress toward understanding two Arabidopsis variegations, immutans (im) and var2 is summarized. Both im and var2 are caused by nuclear recessive mutations and the responsible genes have been cloned and characterized. IMMUTANS functions as a chloroplast terminal oxidase that transfers electrons from the plastoquinol pool to oxygen. It appears to be a versatile electron sink, especially early in chloroplast development, when its function is crucial for carotenoid biosynthesis, and in excess light, when it serves as a 'safety valve'. IM also probably functions in chlororespiration. VAR2 encodes a chloroplast FtsH metalloprotease (termed AtFtsH2). Along with other AtFtsH proteins (AtFtsH1, 5 and 8), it forms complexes in the thylakoid membrane that are probably involved in the process of PSII repair during photoinhibition. A model has been proposed to explain the mechanism of var2 variegation, which suggests that threshold levels of FtsH complexes are required for green sector formation. It is concluded that studies on im and var2 have provided novel insights into nuclear-chloroplast interactions and, especially, into mechanisms of photoprotection.