Пятая колонна в России

Раскрываем тему

В продолжении резонансной перемоги "Про убогий клоачный русский язык в этой стране" (скрины кликабельны) Профессор ВШЭ оправдался за слова об "убогом" русском языке МОСКВА, 29 окт — РИА Новости. Профессор факультета гуманитарных наук Высшей школы экономики Гасан Гусейнов уточнил, что имел в виду,… //  peremogi.livejournal.com

Technological challenges and milestones for writing genomes

Engineering biology with recombinant DNA, broadly called synthetic biology, has progressed tremendously in the last decade, owing to continued industrialization of DNA synthesis, discovery and development of molecular tools and organisms, and increasingly sophisticated modeling and analytic tools. However, we have yet to understand the full potential of engineering biology because of our inability to write and test whole genomes, which we call synthetic genomics. Substantial improvements are needed to reduce the cost and increase the speed and reliability of genetic tools. Here, we identify emerging technologies and improvements to existing methods that will be needed in four major areas to advance synthetic genomics within the next 10 years: genome design, DNA synthesis, genome editing, and chromosome construction (see table). Similar to other large-scale projects for responsible advancement of innovative technologies, such as the Human Genome Project, an international, cross-disciplinary effort consisting of public and private entities will likely yield maximal return on investment and open new avenues of research and biotechnology. The ability to design and write genomes of living cells provides distinctive opportunities to tackle problems that are intractable with current technologies. These transformative technologies could have widespread scientific, social, and economic impacts; thus, their development and adoption require proactive identification of potential pitfalls through ongoing public discussion. Accordingly, the outlook presented here is part of a broader effort by the international Genome Project-write (GP-write) consortium ([ 1 ]) to encourage inclusive conversation among scientists, lawyers, ethicists, educators, environmentalists, other experts, and stakeholders, as well as the general public, and ensure responsible, safe, and coordinated implementation of these new technologies. Synthetic genomics is a relatively new field, and the majority of writing technologies (with the exception of commercial DNA synthesis) are still developed by academic research laboratories. Thus, unlike the reasonably predictable progress of engineering in more established industries, such as semiconductors, prediction of timelines and costs in these nascent fields remains highly speculative (see table). Genome design aims to encode higher-level design criteria into DNA sequences at the chromosome scale. This will require computer-aided design (CAD) technologies to (i) reliably produce desired phenotypes; (ii) maximize the impact of the design, both in terms of experimental feedback and technical feasibility; and (iii) facilitate collaboration by employing standards for handling and exchanging design information. Current synthetic genomic CAD software, such as used for Sc2.0 ([ 2 ]), uses automation and collaboration tools for scaling DNA design from plasmids to entire chromosomes. Currently, estimation of functional effects of edits (e.g., gene deletions) is left to human experts. Looking forward, these tools will need to accurately predict the viability and phenotype of a cell from its genome design. Although simple models are sufficient to handle silent edits (e.g., synonymous editing of coding sequences for watermarking), models of increasing complexity and precision will be necessary to predict how changes to genome sequence affect gene regulation and protein function. Although comprehensive mechanistic models of higher eukaryotes are likely decades away, machine learning approaches could improve phenotype prediction by leveraging the wealth of high-dimensional, high-throughput systems biology data in public databases and generated from genome writing projects, similar to how such techniques have been used to predict protein structures. Leveraging these models, there is a need for experimental design tools to minimize the number of expensive iterations required to obtain successful genome designs. For instance, redesign of the relatively small Mycoplasma genome required four iterations, as well as genomic screens to identify essential genes ([ 3 ]). Larger projects will require many more intermediate stages of construction and supporting experiments. To provide the most valuable feedback and leverage the resulting data in subsequent designs, new algorithms are required to automate design of experiments and selection of appropriate engineering technologies for their implementation (e.g., “write versus edit”). A related, unaddressed need is to ensure compatibility of designed DNA with constraints from downstream synthesis, assembly, delivery, and analytical stages (e.g., software-guided parsing of chromosome sequence into synthesis-compatible fragments, or introduction of designated sequences to facilitate assembly and delivery). Tools will need to be adaptable to anticipate advances in writing technologies. All of these efforts rely on and generate large datasets, which require stewardship to streamline model definition and facilitate sharing of results. //  science.sciencemag.org