黏液霉菌模拟用于绘制暗物质聚集宇宙的地图

公众号：太空梦想家

翻译：小帅

译文

黏液霉菌模拟用于绘制暗物质聚集宇宙的地图

大自然中最不起眼的生物之一的这种行为正帮助天文学家探测宇宙中最大的结构。

这种单细胞生物，被称为黏菌(多头绒泡菌)，为了寻找食物，建立复杂的丝状网络，找到接近最佳的路径来连接不同的位置。在塑造宇宙的过程中，万有引力形成了一个巨大的蜘蛛网结构，它由丝状物构成，将星系和星系团连接在一起，沿着数亿光年长的模糊的桥梁连接在一起。这两个网络有着惊人的相似之处：一个是由生物进化形成的，另一个是由原始的引力形成的。

宇宙网是宇宙的大型骨架，主要由被称为暗物质的神秘物质和气体组成，星系就是在这种物质上形成的。暗物质是看不见的，但它构成了宇宙的大部分物质。1985年在哈佛-史密森天体物理中心进行的红移调查首次暗示了宇宙中存在网状结构。自这些研究以来，在随后的天文观测中，这种丝状结构的规模越来越大。这些细丝形成了宇宙中巨大空洞之间的边界。

但是天文学家很难找到这些难以捉摸的气体束，因为气体太暗，很难探测到。现在，一组研究人员求助于黏液霉菌，帮助他们绘制局部宇宙(距离地球5亿光年以内)的细丝地图，并找到其中的气体。

他们受到黏液霉菌行为的启发，设计了一个计算机算法，并用计算机模拟宇宙中暗物质细丝的生长情况，对其进行了测试。计算机算法类似于一个配方，它精确地告诉计算机要采取什么步骤来解决一个问题。

然后，研究人员将黏液霉菌算法应用到数据中，这些数据包含了由斯隆数字巡天计划（Sloan Digital Sky Survey）绘制的37000个星系的位置，这些星系的距离相当于3亿光年。该算法生成了一张底层宇宙网络结构的三维地图。

然后，他们分析了来自350个类星体（距离数十亿光年，甚至更远）的紫外线，这些类星体被记录在哈勃光谱遗产档案中，这些档案保存着来自NASA哈勃太空望远镜光谱仪的数据。这些遥远的宇宙灯塔是活跃星系的明亮的黑洞供电核心，它们的光照亮了空间和前景的宇宙网络。研究小组在灯丝上的特定位置分析了未被探测到的氢气的吸收特征，并将其印在灯上。这些目标位置远离星系，这使得研究小组能够将气体与宇宙的大规模结构联系起来。

天文学家们在探索难以捉摸的宇宙网络——宇宙的大尺度主干方面变得很有创意。研究人员求助于黏液霉菌，一种在地球上发现的单细胞生物，帮助他们建立局部宇宙(距离地球5亿光年以内)细丝的地图，并找到其中的气体。研究人员根据这种生物的行为设计了一种计算机算法，并将其应用于由斯隆数字巡天计划（Sloan Digital Sky Survey）绘制的37000个星系（黏菌的“食物”）位置的数据中。该算法生成了一张三维地图，显示了底层的宇宙网络错综复杂的丝状网络，即图像中的紫色结构。这三组镶嵌盒显示了一些被“喂”给黏液霉菌和连接它们的丝状结构的个别星系。在三幅插图中，黄色的圆点代表了星系。每个星系的快照旁边都是星系的图像，上面叠加着宇宙网的连线(紫色)。

加州大学圣克鲁斯分校的首席研究员约瑟夫·伯切特（Joseph Burchett）说：“最简单的生命形式竟然能够洞察宇宙中最大规模的结构，这真是令人着迷。通过使用黏液霉菌模拟来寻找宇宙网络细丝的位置，包括那些远离星系的细丝，然后我们可以使用哈勃太空望远镜的档案数据来检测和确定这些看不见的细丝外围的冷气体的密度。几十年来，科学家们已经发现了这种气体的特征，我们已经证实了这种气体构成宇宙网络的理论预期。“

该调查进一步证实了星系间气体密度较大的区域被组织成细丝的研究，该团队发现这些细丝延伸到距离星系超过1000万光年的地方。（这个距离是我们银河系直径的100多倍。)

研究人员在寻找一种方法来可视化宇宙网络结构和之前哈勃光谱研究中探测到的冷气体之间的理论联系时，转向了黏液霉菌模拟。

当时的团队成员、加州大学圣克鲁兹分校的计算媒体科学家奥斯卡•埃莱克（Oskar Elek）在网上发现了塞吉•詹森（Sage Jenson）的作品，后者是一位驻柏林的媒体艺术家。詹森的作品中有令人着迷的艺术视觉效果，展示了黏液霉菌触角状的觅食结构的生长过程。詹森的艺术是基于外部的科学研究，它详细描述了一个模拟黏液霉菌生长的算法。

研究小组注意到黏液霉菌是如何构建复杂的细丝来捕捉新食物的，而重力在塑造宇宙时是如何构建星系和星系团之间的宇宙网络的，两者之间有着惊人的相似之处。

基于这个模拟，埃莱克开发了一个黏液霉菌形成的三维计算机模型来估计宇宙网的丝状结构的位置。

虽然使用黏液霉菌模型仿真确定宇宙最大的结构可能听起来奇怪，科学家们利用计算机模型这些卑微的微生物，以及增加他们在实验室培养皿中，来解决此类复杂问题在大城市找到最有效的交通路线,解决迷宫和精确定位人群疏散路线。“这些都是人类难以解决的难题，更不用说计算机算法了。”埃莱克说。

“你几乎可以从斯隆的数据中看到，特别是在局部宇宙的星系地图中，细丝应该在哪里。”伯切特解释道，”黏液霉菌模型非常符合这种直觉。你知道应该在那里的结构是突然被计算机算法发现的。对于我们的研究来说，没有其他已知的方法可以很好地解决这个问题。”

研究人员表示，在如此大规模的星系调查中，很难设计出一种可靠的算法来寻找灯丝。“所以，看到虚拟黏液霉菌在短短几分钟内给你一个非常接近的近似值是非常令人惊讶的。”埃莱克解释说，“你真的可以看着它成长。”只是比较一下，在培养皿中培养微生物需要几天时间。黏液霉菌实际上有一种非常特殊的智能来解决这个空间任务。毕竟，这对它的生存至关重要。

研究小组的论文将发表在《天体物理学杂志通讯》上。

哈勃太空望远镜是NASA和ESA的国际合作项目。位于马里兰州格林贝尔特的美国宇航局戈达德太空飞行中心负责管理这架望远镜。位于马里兰州巴尔的摩市的太空望远镜科学研究所（STScI）负责哈勃的科学运作。STScI是由位于华盛顿特区的天文学研究大学协会为NASA运作的。

原文

Slime Mold Simulations Used to Map Dark Matter Holding Universe Together

The behavior of one of nature's humblest creatures is helping astronomers probe the largest structures in the universe.

The single-cell organism, known as slime mold (Physarum polycephalum), builds complex filamentary networks in search of food, finding near-optimal pathways to connect different locations. In shaping the universe, gravity builds a vast cobweb structure of filaments tying galaxies and clusters of galaxies together along faint bridges hundreds of millions of light-years long. There is an uncanny resemblance between the two networks: one crafted by biological evolution, and the other by the primordial force of gravity.

The cosmic web is the large-scale backbone of the cosmos, consisting primarily of the mysterious substance known as dark matter and laced with gas, upon which galaxies are built. Dark matter cannot be seen, but it makes up the bulk of the universe's material. The existence of a web-like structure to the universe was first hinted at in the 1985 Redshift Survey conducted at the Harvard-Smithsonian Center for Astrophysics. Since those studies, the grand scale of this filamentary structure has grown in subsequent sky surveys. The filaments form the boundaries between large voids in the universe.

But astronomers have had a difficult time finding these elusive strands, because the gas is so dim it is hard to detect. Now a team of researchers has turned to slime mold to help them build a map of the filaments in the local universe (within 500 million light-years from Earth) and find the gas within them.

They designed a computer algorithm, inspired by slime-mold behavior, and tested it against a computer simulation of the growth of dark matter filaments in the universe. A computer algorithm is similar to a recipe that tells a computer precisely what steps to take to solve a problem.

The researchers then applied the slime mold algorithm to data containing the locations of 37,000 galaxies mapped by the Sloan Digital Sky Survey at distances corresponding to 300 million light-years. The algorithm produced a three-dimensional map of the underlying cosmic web structure.

They then analyzed the ultraviolet light from 350 quasars (at much farther distances of billions of light-years) cataloged in the Hubble Spectroscopic Legacy Archive, which holds the data from NASA's Hubble Space Telescope's spectrographs. These distant cosmic flashlights are the brilliant black-hole-powered cores of active galaxies, whose light shines across space and through the foreground cosmic web. Imprinted on that light was the telltale absorption signature of otherwise undetected hydrogen gas that the team analyzed at specific points along the filaments. These target locations are far from the galaxies, which allowed the research team to link the gas to the universe's large-scale structure.

Astronomers have gotten creative in trying to trace the elusive cosmic web, the large-scale backbone of the cosmos. Researchers turned to slime mold, a single-cell organism found on Earth, to help them build a map of the filaments in the local universe (within 500 million light-years from Earth) and find the gas within them. The researchers designed a computer algorithm inspired by the organism's behavior and applied it to data containing the positions of 37,000 galaxies ("food" for the slime mold) mapped by the Sloan Digital Sky Survey. The algorithm produced a three-dimensional map of the underlying cosmic web's intricate filamentary network, the purple structure in the image. The three sets of inset boxes show some of those individual galaxies that were "fed" to the slime mold and the filamentary structure connecting them. The galaxies are represented by the yellow dots in three of the inset images. Next to each galaxy snapshot is an image of the galaxies with the cosmic web's connecting strands (purple) superimposed on them.

"It's really fascinating that one of the simplest forms of life actually enables insight into the very largest-scale structures in the universe," said lead researcher Joseph Burchett of the University of California (UC), Santa Cruz. "By using the slime-mold simulation to find the location of the cosmic web filaments, including those far from galaxies, we could then use the Hubble Space Telescope's archival data to detect and determine the density of the cool gas on the very outskirts of those invisible filaments. Scientists have detected signatures of this gas for several decades, and we have proven the theoretical expectation that this gas comprises the cosmic web."

The survey further validates research that denser regions of intergalactic gas is organized into filaments that the team found stretches over 10 million light-years from galaxies. (That distance is more than 100 times the diameter of our Milky Way galaxy.)

The researchers turned to slime mold simulations when they were searching for a way to visualize the theorized connection between the cosmic web structure and the cool gas detected in previous Hubble spectroscopic studies.

Then team member Oskar Elek, a computational media scientist at UC Santa Cruz, discovered online the work of Sage Jenson, a Berlin-based media artist. Among Jenson's works were mesmerizing artistic visualizations showing the growth of a slime mold's tentacle-like network of food-seeking structures. Jenson's art was based on outside scientific research, which detailed an algorithm for simulating the growth of slime mold.

The research team noted a striking similarity between how the slime mold builds complex filaments to capture new food, and how gravity, in shaping the universe, constructs the cosmic web strands between galaxies and galaxy clusters.

Based on the simulation, Elek developed a three-dimensional computer model of the buildup of slime mold to estimate the location of the cosmic web's filamentary structure.

Although using a slime-mold-inspired simulation to pinpoint the universe's largest structures may sound bizarre at first, scientists have used computer models of these humble microorganisms, as well as grown them in petri dishes in a lab, to solve such complex problems as finding the most efficient traffic routes in large cities, solving mazes and pinpointing crowd evacuation routes. "These are hard problems to solve for a human, let alone a computer algorithm," Elek said.

"You can almost see, especially in the map of galaxies in the local universe from the Sloan data, where the filaments should be," Burchett explained. "The slime-mold model fits that intuition impressively. The structure that you know should be there is all of a sudden found by the computer algorithm. There was no other known method that was well suited to this problem for our research."

The researchers say that it is very difficult to design a reliable algorithm for finding the filaments in such a large survey of galaxies. "So it's quite amazing to see that the virtual slime mold gives you a very close approximation in just minutes," Elek explained. "You can literally watch it grow." Just for comparison, growing the organism in a petri dish takes days. Slime mold actually has a very special kind of intelligence for solving this one spatial task. After all, it's critical to its survival.

The team's paper will appear in The Astrophysical Journal Letters.

The Hubble Space Telescope is a project of international cooperation between NASA and ESA (European Space Agency). NASA's Goddard Space Flight Center in Greenbelt, Maryland, manages the telescope. The Space Telescope Science Institute (STScI) in Baltimore, Maryland, conducts Hubble science operations. STScI is operated for NASA by the Association of Universities for Research in Astronomy in Washington, D.C.

每日一图

公众号：太空梦想家

原文链接：https://www.nasa.gov/feature/goddard/2020/slime-mold-simulations-used-to-map-dark-matter-holding-universe-together

注：所有视频和图片均来自NASA