Refining CycleGAN with attention mechanisms and age-Aware training for realistic Deepfakes.

In the evolving landscape of deep learning technologies, the emergence of Deepfakes and synthetic media is becoming increasingly prominent within digital media production. This research addresses the limitations inherent in existing face image generation algorithms based on Generative Adversarial Networks (GAN), particularly the challenges of domain irrelevancy and inadequate facial detail representation. The study introduces an enhanced face image generation algorithm, aiming to refine the CycleGAN framework. The enhancement involves a two-fold strategy: firstly, the generator's architecture is refined through the integration of an attention mechanism and adaptive residual blocks, enabling the extraction of more nuanced facial features. Secondly, the discriminator's accuracy in distinguishing real from synthetic images is improved by incorporating a relative loss concept into the loss function. Additionally, this study presents a novel model training approach that incorporates age constraints, thereby mitigating the effects of age variations on the synthesized images. The effectiveness of the proposed algorithm is empirically validated through comparative analysis with existing methodologies, utilizing the CelebA dataset. The results demonstrate that the proposed algorithm significantly enhances the realism of generated face images, outperforming current methods in terms of Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM), while also achieving notable improvements in subjective visual quality. The implementation of this advanced method is anticipated to substantially elevate the efficiency and quality of digital media production, contributing positively to the broader field of digital media creation.

Growth Efficiency of Chlorella sorokiniana in Synthetic Media and Unsterilized Domestic Wastewater.

Incorporating a variety of microalgae into wastewater treatment is considered an economically viable and environmentally sound strategy. The present work assessed the growth characteristics of Chlorella sorokiniana during cultivation in balanced synthetic media and domestic wastewater. Increasing the NH4+-N concentration to 360 mg L-1 and adding extra PO43--P and SO42--S (up to 80 and 36 mg L-1, respectively) contributed to an increase in the total biomass levels (5.7-5.9 g L-1) during the cultivation of C. sorokiniana in synthetic media. Under these conditions, the maximum concentrations of chlorophylls and carotenoids were 180 ± 7.5 and 26 ± 1.4 mg L-1, respectively. Furthermore, when studying three types of domestic wastewaters, it was noted that only one wastewater contributed to the productive growth of C. sorokiniana, but all wastewaters stimulated an increased accumulation of protein. Finally, the alga, when growing in optimal unsterilized wastewater, showed a maximum specific growth rate of 0.73 day-1, a biomass productivity of 0.21 g L-1 day-1, and 100% NH4+-N removal. These results demonstrate that the tested alga actively adapts to changes in the composition of the growth medium and accumulates high levels of protein in systems with poor-quality water.

Transition from synthetic to alternative media for microalgae cultivation: A critical review.

In recent decades, microalgae have drawn attention as a most feasible alternative and sustainable feedstock for biofuel production. However, laboratory-scale and pilot-scale studies revealed that producing only biofuels through the microalgal route is economically unfeasible. The high cost of synthetic media is one concern, and low-cost alternative cultivation media would replace synthetic media to culture microalgae for economic benefit. This paper critically consolidated the advantages of alternative media over synthetic media for microalgae cultivation. A comparative analysis of the compositions of synthetic and alternative media was made to evaluate the potential use of alternative media in microalgae cultivation. Investigations on microalgae cultivation using alternative media derived from different waste materials, such as domestic, farm, agricultural, industrial, etc., are highlighted. Vermiwash is another alternative media that contains essential micro and macronutrients required for the cultivation of microalgae. Two prime techniques, such as mix-waste culture media and recycling culture media, may provide more economic benefit for the large-scale production of microalgae.

Regulatory mechanism of trichothecene biosynthesis in Fusarium graminearum.

Among the genes involved in the biosynthesis of trichothecene (Tri genes), Tri6 and Tri10 encode a transcription factor with unique Cys2His2 zinc finger domains and a regulatory protein with no consensus DNA-binding sequences, respectively. Although various chemical factors, such as nitrogen nutrients, medium pH, and certain oligosaccharides, are known to influence trichothecene biosynthesis in Fusarium graminearum, the transcriptional regulatory mechanism of Tri6 and Tri10 genes is poorly understood. Particularly, culture medium pH is a major regulator in trichothecene biosynthesis in F. graminearum, but it is susceptible to metabolic changes posed by nutritional and genetic factors. Hence, appropriate precautions should be considered to minimize the indirect influence of pH on the secondary metabolism while studying the roles of nutritional and genetic factors on trichothecene biosynthesis regulation. Additionally, it is noteworthy that the structural changes of the trichothecene gene cluster core region exert considerable influence over the normal regulation of Tri gene expression. In this perspective paper, we consider a revision of our current understanding of the regulatory mechanism of trichothecene biosynthesis in F. graminearum and share our idea toward establishing a regulatory model of Tri6 and Tri10 transcription.

Protecting world leaders against deep fakes using facial, gestural, and vocal mannerisms.

Since their emergence a few years ago, artificial intelligence (AI)-synthesized media-so-called deep fakes-have dramatically increased in quality, sophistication, and ease of generation. Deep fakes have been weaponized for use in nonconsensual pornography, large-scale fraud, and disinformation campaigns. Of particular concern is how deep fakes will be weaponized against world leaders during election cycles or times of armed conflict. We describe an identity-based approach for protecting world leaders from deep-fake imposters. Trained on several hours of authentic video, this approach captures distinct facial, gestural, and vocal mannerisms that we show can distinguish a world leader from an impersonator or deep-fake imposter.

Development of a novel defined minimal medium for Gluconobacter oxydans 621H by systematic investigation of metabolic demands.

BACKGROUND: Historically, complex media are used for the cultivation of Gluconobacter oxydans in industry and research. Using complex media has different drawbacks like higher costs for downstream processing and significant variations in fermentation performances. Synthetic media can overcome those drawbacks, lead to reproducible fermentation performances. However, the development of a synthetic medium is time and labour consuming. Detailed knowledge about auxotrophies and metabolic requirements of G. oxydans is necessary. In this work, we use a systematic approach applying the in-house developed µRAMOS technology to identify auxotrophies and develop a defined minimal medium for cultivation of G. oxydans fdh, improving the production process of the natural sweetener 5-ketofructose. RESULTS: A rich, defined synthetic medium, consisting of 48 components, including vitamins, amino acids and trace elements, was used as a basis for medium development. In a comprehensive series of experiments, component groups and single media components were individually omitted from or supplemented to the medium and analysed regarding their performance. Main components like salts and trace elements were necessary for the growth of G. oxydans fdh, whereas nucleotides were shown to be non-essential. Moreover, results indicated that the amino acids isoleucine, glutamate and glycine and the vitamins nicotinic acid, pantothenic acid and p-aminobenzoic acid are necessary for the growth of G. oxydans fdh. The glutamate concentration was increased three-fold, functioning as a precursor for amino acid synthesis. Finally, a defined minimal medium called 'Gluconobacter minimal medium' was developed. The performance of this medium was tested in comparison with commonly used media for Gluconobacter. Similar/competitive results regarding cultivation time, yield and productivity were obtained. Moreover, the application of the medium in a fed-batch fermentation process was successfully demonstrated. CONCLUSION: The systematic investigation of a wide range of media components allowed the successful development of the Gluconobacter minimal medium. This chemically defined medium contains only 14 ingredients, customised for the cultivation of G. oxydans fdh and 5-ketofructose production. This enables a more straightforward process development regarding upstream and downstream processing. Moreover, metabolic demands of G. oxydans were identified, which further can be used in media or strain development for different processes.

Deepfakes and Democracy (Theory): How Synthetic Audio-Visual Media for Disinformation and Hate Speech Threaten Core Democratic Functions.

Observers fear that deepfakes will shake the very foundations of democracy. Notwithstanding, in-depth scholarly analyses of deepfakes' political impact are rare, and do not consider theories of democracy. This contribution helps close this research gap, drawing on Warren's problem-oriented democracy theory, as well as theories of deliberative democracy and contributions on the role of trust in democracies. I identify three core functions of democratic systems and their normative foundations, namely empowered inclusion, collective agenda and will formation (supported by deliberation), and collective decision-making. Based on a literature and media analysis, I systematize different types of deepfakes serving either disinformation or hate speech and outline how they weaken core democratic functions and norms: Deepfakes impede citizens' empowered inclusion in debates and decisions that affect them, e.g. by hampering efforts to hold political representatives accountable or further marginalizing certain societal groups such as women or ethnic minorities. Deepfakes also undermine collective agenda and will formation by threatening the epistemic quality of deliberation as well as citizens' mutual empathy and respect. This culminates in a decreased legitimacy of collective decisions taken, which is additionally threatened by pervasive (but mostly speculative) fears of deepfake election manipulation. My analysis has implications for (future) governance efforts addressing deepfakes. Such efforts are increasing, e.g. on the part of social media platforms, but also (supra-)national regulatory bodies.

Problematising reality: the promises and perils of synthetic media.

This commentary article focuses on the emergence of synthetic media-computer-generated content that is created by employing Artificial Intelligence (AI) technologies. It discusses three of the most notable current forms of this emerging form of content: deepfakes, virtual influencers, and augmented and virtual reality (collectively known as extended reality). Their key features are introduced, and the main challenges and opportunities associated with the technologies are analysed. In all cases, a crucial change is underway: reality (or, at least, the perception thereof) is seen as increasingly less stable, and potential for manipulation is on the rise. In fact, it transpires that personalisation of (perceived) reality is the likely outcome, with increasing societal fragmentation as a result. Mediatisation is used as a broad-ranging metatheory that explains the permeation by media of everyday affairs to explain the degree of impact that synthetic media have on the society. In this context, it is suggested that we search for new and alternative criteria for reality that would be capable of accounting for the changing nature of agency and impact in today's world.

A Detection Method of Operated Fake-Images Using Robust Hashing.

SNS providers are known to carry out the recompression and resizing of uploaded images, but most conventional methods for detecting fake images/tampered images are not robust enough against such operations. In this paper, we propose a novel method for detecting fake images, including distortion caused by image operations such as image compression and resizing. We select a robust hashing method, which retrieves images similar to a query image, for fake-image/tampered-image detection, and hash values extracted from both reference and query images are used to robustly detect fake-images for the first time. If there is an original hash code from a reference image for comparison, the proposed method can more robustly detect fake images than conventional methods. One of the practical applications of this method is to monitor images, including synthetic ones sold by a company. In experiments, the proposed fake-image detection is demonstrated to outperform state-of-the-art methods under the use of various datasets including fake images generated with GANs.

Optimizing cultivation of Cordyceps militaris for fast growth and cordycepin overproduction using rational design of synthetic media.

Cordyceps militaris is an entomopathogenic fungus which is often used in Asia as a traditional medicine developed from age-old wisdom. Presently, cordycepin from C. militaris is a great interest in medicinal applications. However, cellular growth of C. militaris and the association with cordycepin production remain poorly understood. To explore the metabolism of C. militaris as potential cell factories in medical and biotechnology applications, this study developed a high-quality genome-scale metabolic model of C. militaris, iNR1329, based on its genomic content and physiological data. The model included a total of 1329 genes, 1821 biochemical reactions, and 1171 metabolites among 4 different cellular compartments. Its in silico growth simulation results agreed well with experimental data on different carbon sources. iNR1329 was further used for optimizing the growth and cordycepin overproduction using a novel approach, POPCORN, for rational design of synthetic media. In addition to the high-quality GEM iNR1329, the presented POPCORN approach was successfully used to rationally design an optimal synthetic medium with C:N ratio of 8:1 for enhancing 3.5-fold increase in cordycepin production. This study thus provides a novel insight into C. militaris physiology and highlights a potential GEM-driven method for synthetic media design and metabolic engineering application. The iNR1329 and the POPCORN approach are available at the GitHub repository: https://github.com/sysbiomics/Cordyceps_militaris-GEM.

Vitamin requirements and biosynthesis in Saccharomyces cerevisiae.

Chemically defined media for yeast cultivation (CDMY) were developed to support fast growth, experimental reproducibility, and quantitative analysis of growth rates and biomass yields. In addition to mineral salts and a carbon substrate, popular CDMYs contain seven to nine B-group vitamins, which are either enzyme cofactors or precursors for their synthesis. Despite the widespread use of CDMY in fundamental and applied yeast research, the relation of their design and composition to the actual vitamin requirements of yeasts has not been subjected to critical review since their first development in the 1940s. Vitamins are formally defined as essential organic molecules that cannot be synthesized by an organism. In yeast physiology, use of the term "vitamin" is primarily based on essentiality for humans, but the genome of the Saccharomyces cerevisiae reference strain S288C harbours most of the structural genes required for synthesis of the vitamins included in popular CDMY. Here, we review the biochemistry and genetics of the biosynthesis of these compounds by S. cerevisiae and, based on a comparative genomics analysis, assess the diversity within the Saccharomyces genus with respect to vitamin prototrophy.

Advances in yeast alcoholic fermentations for the production of bioethanol, beer and wine.

Yeasts have a long-standing relationship with humankind that has widened in recent years to encompass production of diverse foods, beverages, fuels and medicines. Here, key advances in the field of yeast fermentation applied to alcohol production, which represents the predominant product of industrial biotechnology, will be presented. More specifically, we have selected industries focused in producing bioethanol, beer and wine. In these bioprocesses, yeasts from the genus Saccharomyces are still the main players, with Saccharomyces cerevisiae recognized as the preeminent industrial ethanologen. However, the growing demand for new products has opened the door to diverse yeasts, including non-Saccharomyces strains. Furthermore, the development of synthetic media that successfully simulate industrial fermentation medium will be discussed along with a general overview of yeast fermentation modeling.

Comparison of Synthetic Media Designed for Expansion of Adipose-Derived Mesenchymal Stromal Cells.

Mesenchymal stromal cells (MSCs) are multipotent cells that can differentiate into various cell types, such as osteoblasts, myocytes, and adipocytes. This characteristic makes the cells a useful tool in developing new therapies for a number of common maladies and diseases. The utilization of animal-derived growth serum, such as fetal bovine serum (FBS), for the expansion of MSCs has traditionally been used for cell culture. However, in clinical applications, animal-derived products present limitations and safety concerns for the recipient, as exposure to animal (xeno-) antigens and infectious agents is possible. Multiple synthetic, xeno-free media have been developed to combat these limitations of animal-derived growth serum and have the potential to be used in ex vivo MSC expansion for clinical use. The goal of this study was to determine if xeno-free media are adequate to significantly and efficiently expand MSCs derived from adipose tissue. MSCs were cultured in both standard FBS-containing as well as xeno-free media. The media were compared for cell yield, viability, and phenotypic expression via flow cytometry and directed differentiation. The xeno-free media that were tested were StemMACS MSC Expansion Media (Miltenyi Biotec, Bergisch Gladbach, Germany), PLTMax Human Platelet Lysate (Sigma-Aldrich, St. Louis, MO, USA), and MesenCult-hPL media (Stemcell Technologies, Vancouver, BC, Canada). All xeno-free media showed promise as a feasible replacement for animal-derived growth serums. The xeno-free media expanded MSCs more quickly than the FBS-containing medium and also showed great similarity in cell viability and phenotypic expression. In fact, each xeno-free media produced a greater viable cell yield than the standard FBS-containing medium.