26 research outputs found
Concise synthesis of vinylheterocycles through beta-elimination under solventless phase transfer catalysis conditions
No full textVarious vinylheterocycles compounds have been prepared in excellent yields through betaelimination of the corresponding sulfonate esters with 50% aq NaOH under phase transfer catalysis conditions without organic solvent. The new approach provides an economic and environmentally friendly solution to removal of hazardous bases as well as toxic and expensive dipolar aprotic solvents
Compression strategies and space-conscious representations for deep neural networks
No full textRecent advances in deep learning have made available large, powerful convolutional neural networks (CNN) with state-of-the-art performance in several real-world applications. Unfortunately, these large-sized models have millions of parameters, thus they are not deployable on resource-limited platforms (e.g., where RAM is limited). Compression of CNNs becomes therefore a critical problem to achieve memory-efficient and possibly computationally faster model representations. In this paper, we investigate the impact of lossy compression of CNNs by weight pruning and quantization, and lossless weight matrix representations based on source coding. We tested several combinations of these techniques on four benchmark datasets for classification and regression problems, achieving compression rates up to 165 times, while preserving or improving the model performance
Endophytes isolated from maize Lombard landraces: new perspectives for the control of Fusarium verticillioides
No full textMaize is one of the most important crops in our country. However, in the past years yields have gradually decreased due to reduced cultivated areas and climate change. GEMMA project aims to study the endophytic bacterial community of landraces and their relationship with Fusarium verticillioides, one of the most widespread toxicogenic fungal pathogens in Lombardy region. Recent studies show how endophytes have abilities to inhibit the development of pathogens. Over the past three years four traditional landraces (Nero Spinoso Valcamonica, Spinato Gandino, Rostrato Rosso Rovetta, Fiorine Clusone, preserved at CREA Bergamo Genebank) and a pure line (B73) were grown in different locations (Landriano, Bergamo, Verderio, Carvico) with low-input farming techniques. Embryo of seeds harvested in 2019 allowed the isolation of 96 bacterial strains, mostly Firmicutes. In vitro antifungal assays against F. verticillioides allowed the selection of 12 strains with inhibition percentage higher than 50% for in vivo test. Only 2 isolates (Bacillus and Psychrobacillus) significantly reduced fungal infection in vivo. In 2021, experimental inoculations with F. verticillioides conidia were carried out in field during flowering. Values of infected seeds/ear and % infected area/ear show that Spinato di Gandino is the most susceptible variety while Nero Spinoso is confirmed as being the least susceptible.
The study of microbiota and resistance traits from landraces shows potential for future organic maize production where biocontrol agents and resistant genotypes can contribute to a better, and more sustainable, response to stresses that will allow a higher maize yield even in the scenario of climate change
PGRFA Management of Outcrossing Plants Propagated by Seed: From On-Farm to Ex Situ Conservation and Some Italian Maize Case Studies
No full textIn this review, the main issues related to the conservation and valorization of Plant Genetic Resources for Food and Agriculture (PGRFA) will be primarily addressed. The conservation of PGRFA concerning outcrossing plants poses a significant challenge. For this reason, this review will cover the key challenges related to all stages, starting from in situ sampling, collection in the germplasm bank, and conservative reproductive methods. Integrated approaches involving the combined use of classical and molecular techniques will be described for the characterization of accessions. Within this framework, some successful Italian case studies focused on maize will be reported as well
Reproducing the Sparse Huffman Address Map Compression for Deep Neural Networks
No full textDeploying large convolutional neural networks (CNNs) on limited-resource devices is still an open challenge in the big data era. To deal with this challenge, a synergistic composition of network compression algorithms and compact storage of the compressed network has been recently presented, substantially preserving model accuracy. The proposed implementation, which we describe in this paper, offers different compression schemes (pruning, two types of weight quantization, and their combinations) and two compact representations: the Huffman Address Map compression (HAM), and its sparse version sHAM. Taken as input a model, trained for a given classification or regression problem (as well as the dataset employed, which is necessary for the fine-tuning of weights after network compression), the procedure returns the corresponding compressed model. Our publicly available implementation provides the source code, two pre-trained CNN models (retrieved from third-party repositories referring to well-established literature), and four datasets. This implementation includes detailed instructions to execute the scripts and reproduce the obtained results, in terms of the figures and tables included in the original paper
Preliminary Evidence of a Horizontal Transfer of Paramutation Phenomenon at the pl1 Gene in Maize (Zea mays L.)
No full textParamutation, a specific epigenetic phenomenon first identified in Zea mays by Alexander Brink in the 1950s, has since been observed in different plant and animal species. What sets paramutation apart from other gene silencing processes is its ability for one silenced allele (referred to as paramutagenic) to silence another allele (paramutable) in trans. The resultant silenced allele (paramutated) remains stable across generations, even after separating from the paramutagenic allele, and acquires paramutagenic properties itself. This hereditary behavior facilitates the rapid dissemination of specific gene expression patterns or phenotypes within populations, disrupting the Hardy-Weinberg equilibrium even without other evolutionary pressures. Despite extensive research, the molecular mechanisms underlying paramutation are still not fully understood, although numerous studies suggest the involvement of RNA-mediated changes in DNA methylation and in the chromatin structure of silenced genes. In this paper, we report preliminary evidence regarding horizontal paramutation transfer at the pl1 (purple plant1) regulatory gene involved in the accumulation of anthocyanin in several plant tissues such as tassel, kernel, and cob. A paramutated pl1 ' allele arose spontaneously in a pl1 population, and in this study, we found evidence of silencing events spatially associated in the field, suggesting a possible horizontal transfer of silencing among nearby plants
PGRFA management of outcrossing plants propagated by seed: from on-farm to ex situ conservation and some Italian maize case studies
No full textIn this review, the main issues related to the conservation and valorization of Plant Genetic Resources for Food and Agriculture (PGRFA) will be primarily addressed. The conservation of PGRFA concerning outcrossing plants poses a significant challenge. For this reason, this review will cover the key challenges related to all stages, starting from in situ sampling, collection in the germplasm bank, and conservative reproductive methods. Integrated approaches involving the combined use of classical and molecular techniques will be described for the characterization of accessions. Within this framework, some successful Italian case studies focused on maize will be reported as well
Low-Phytate Grains to Enhance Phosphorus Sustainability in Agriculture: Chasing Drought Stress in lpa1-1 Mutant
No full textPhytic acid (PA) is an anti-nutritional factor for monogastrics and contributes to phosphorus pollution. The low phytic acid (lpa) trait can provide several benefits to the nutritional quality of foods/feeds and to environmental sustainability. In maize, four lpa1 mutants have been isolated, and lpa1-1 is the most promising. Nevertheless, these mutations are frequently accompanied by many negative pleiotropic effects affecting plant performance. One of these is a greater susceptibility to drought stress, probably caused by an alteration in the root system. In this work, we set up an experiment in hydroponics and two in mesocosms, where pots were built using transparent PVC sheets to better access the roots. The results suggested that neither root architecture nor root depth are limiting factors in mutant plants. In hydroponics, the dry weight of the mutant and the root area per unit of length were twice that of B73. However, lpa1-1 exhibited a reduced efficiency of photosystem II (Fv/Fm, 0.810 vs. 0.800) and a reduced leaf temperature (-0.5 degrees C compared to wild-type), probably due to increased water loss. Furthermore, molecular analysis performed on genes involved in root development (rtcs, rtcl, rum1, and BIGE1) revealed the abundance of rtcs transcripts in the mutant, suggesting an alteration in auxin polar transport
Improving biodiesel sustainability by double-cropping oilseed and oil-less seed cake recovery: A field approach in northern Italy
No full textThis work proposes a new bioenergy supply chain based on double-cropping oilseed (Camelina sativa and Ricinus communis) in northern Italy, to produce biodiesel from oil and biogas from co-products, improving biodiesel sustainability according to the Renewable Energy Directive (RED II). To achieve this, the crops were cultivated on silty-sandy soil during two seasons. Rainfall was below the annual average during the first season (460 mm) and regular during the second (713 mm); irrigation was provided during summer. After milling, oil and seed cakes were chemically characterized. Subsequently, the potential bio-methane production (BMP) of seed cakes was evaluated. Field trials resulted in average seed yields (values reported as average ± std dev) of 6,948 ± 330 kg TS ha−1 year−1. Chemical characteristics of oil were considered compatible with biodiesel synthesis, while seed cakes BMP (values reported as average ± std dev), i.e., 239 ± 10 NmlCH4 g TS−1 and 337 ± 15 NmlCH4 g TS−1 for C. sativa and R. communis, were comparable to maize. The impact analysis was based on life cycle assessment (LCA); comparison with fossil fuels employed the Fossil Fuel Compensator value, established by RED II. Calculations are based on the reference impact of transport in the European Union (94 gCO2eq MJ−1). LCA showed a 75% GHG reduction vs. petrol diesel; moreover, using nitrification inhibitors (NIs) to reduce N2O emissions, the reduction could reach 81%. Economic viability is based on double cropping and dual energy production; the costs of NIs and cake processing for biogas being negligible
