Exploring Species in Wolfram Language

Hi. Hello everyone. My name is Keiko Hirama, I'm a biology developer in Wolfram Alpha Scientific Content group. Today I would like to share how we can use the new biology data collection to explore species taxonomy information in the Wolframar language. So we recently introduced a new entity data collections called Taxonomic species which is based on NCBI taxonomy database providing hierarchical relational information for over 2 million taxa, including not only animals and plants, but also fungus and other microorganisms as well. And we have curated data set that includes biological, geological and ecological information as well. And data is restructured in order to support advanced queries and analysis of diverse organisms. And one of the key new features about this data set is that we support folk taxonomy group. Fish, otters, monkeys are so called paraphyletic taxonomy group, meaning there is no single taxon to represent disease groups in taxonomy taxonomic 3 because their species belong to different branches in taxonomic tree. So for example some lobe finned fish in this illustration you can see that they belong to the same taxonomic branch as mammals, birds or amphibians, while ray finned fish or cartilaginous fish they have on different branches represented in the tree. So these are so called folk taxonomy group and this is now represented as an entity species phyletic group. So as a result we can actually query what taxomic species belong to this fish group. So as you can see here you can see the list of species belong to fish phyletic group. Another key feature is that this taxonomic species entity type is hierarchically organized. So on top of the here hierarchy, just like taxonomic tree, taxonomic species on the top and then below we have entity types to represent archaea, bacteria, virus, eukaryotes and even below we have the entity types to represent fungus, plant, animal species. And there's a new entity type syntax that you can query. What are the child types of given entity type like taxonomic species, or you can ask for the parent type of given entity type and this top in here this taxonomic species entity type includes taxonomic information like class, family, family list, those kind of information as well as common properties that are shared across all species domain like conservation status, some genomic information like nucleotide sequence length observation locations, biome information or height length lifespan information. Those are shared across all species domain. So included in this taxonomic species level properties. And child types like animal inherit properties of their parents. So including our taxonomic information properties to describe taxonomic information, but also includes some unique properties that are specific to animals. So those are described here. For example, some properties describe reproductive features, behavioral features, information about speed and so on. So those are represented in the properties of animals specifically. And overall we have over 400 properties to describe various species. Especially we have good coverage on plants and animals. But we are trying to expand this data set to include more information on fungus and other microorganisms as well. And for taxonomic properties, we introduced new taxonomic sequence property to provide more detailed interlinked level taxa information. So previous species data we only had taxa to represent a primary taxonomic sequence, meaning just kingdom phylum to species. But in this new taxonomic species data set we have taxa to represent interrank level taxa like superphylum subphylum subclass. So those are represented in also in this data set. So as a result we can actually get more detailed taxonomic sequence information you can query, including intra rank level taxo. And we recently introduced this new resource function called taxonomy graph that you can actually plug in any taxonomic species entity of your interest and you can actually generate nice taxonomic tree graph shown like here. And this taxonomy graph function takes any options from the graph general graph function in the Wolfram language. So you can actually customize the style of this graph as well. And information about taxonomic rank itself is also available. So if you're not sure about what are the sequence about the taxonomic ranks, you can ask for what is the parent or child of phylum taxonomic level. And you can ask for that as well. And using this all the taxonomic rank information, you can also compute a list of species belong to a given taxonomic branch. So for example, here you can ask all species belong to the elephant family or even for higher taxa like mammals, amphibians, birds, you can qualify for the list or count of species belong to this group. Or even you can ask all the genus level species belong to a given taxonomic branch. Another thing we can do is access taxonomic authority information. So some taxonomy can be originally reported, but later it can be merged, renamed or even reclassified, placed in a different place in the taxonomic branch. And that kind of information is also available. And here, this here for example, you can ask what is the when first originally that particular organism is taxonomically classified or when the latest ear that given taxonomy is reclassified modified. So that kind of information is also available. So this date histogram is showing when first these 2 million taxa are first reported. And you can see some species are reported as early as like mid-1700s, some are reported in 1900s and even yeah, nowadays hundreds of new species are reported every year. And we can also explore biological properties of species as well. And great thing about this new taxonomic species data structure is that you can Actually aggregate easily values for folk taxonomy group. So for monkeys, another example of folk taxonomy group. And you can query for example weight of all monkey species. And you can get a nice data set like this listing species and their weight. And you can see that in this box with culture that, you know, mean median weight of monkey species is about like 5 kilogram, but so you can see that their weight greatly varies by different type of different species of monkeys. So that kind of analysis can be done easily with this data structure. Another thing you can do is ask for genome information. We have nucleotide sequence lengths and chromosome information from ncbi. So for example, this is specifically asking five species that have a largest genome. And here lungfish apparently have a nucleotide sequence length of 35 gigabase pairs. And next one is showing like five species that have largest number of chromosomes. And surprisingly this phalan particular phalan species half 480 chromosomes. You can also get genome assembly information for all million species. You can ask for this reference sequence assembly accession number that gives you this nice external identifier object that actually links to the NCBI website where you can also download retrieve the genome sequence, for example in the fast format you can download the file or see ARM genome sequence. And there's a resource function called phylogenetic tree plot. And you can actually plug in your sequence you downloaded and you can also create a nice phylogenetic tree as well based on this sequence analysis. And another thing we can do is look up animals by observation locations, not by country or ocean, but we also support various GEO entities like iround, lake park, volcano or even waterfall. So any type of GEO entities can be used to ask the list of species that are found in the particular location. So here for example, this is specifically querying what are the species of squirrels that are found in Zion National Park. And you get nice pictures here. And next one is showing the mammals found in particular ocean balisi or reversely you can also ask what are the like for example oceans that particular species found. And you get the list of ocean entities object and then you can plug it in the geographics function and then you can get a nice image map illustrating where these dugong species are found. And for migratory animals like osprey, for example, we can ask for the observation locations by date. So this returns the association of date and geopoints so that you can create animation like this to explore the seasonal like migratory pattern of an animal over the course of year, so you can explore how they migrate over the course of a. And of course we can just simply Ask when the like when the migration starts for particular bird or any like for example butterflies in spring. And for that can be queried here as well. Next example, I would like to show how we can do some ecological analysis. We have conservation status information for over 70,000 species, including latest extinct species information. So here, for example, I specifically ask what are the mammals in marine biome that are recently extinct. Or you can color countries by the number of endangered mammal species found using this geo region value plot. So darker color representing the countries where the largest number of endangered species are reported. We can also compare the habitats of endangered species. So we can ask for biomes that endangered species belong. And for each terrestrial biome also represented as an entity. And it has a nice polygon that you can illustrate in the map, it's color coded, for example red is copper biome, brown is desert or dune biome. And by comparing these two, this is showing the endangered mammals reported in the terrestrial biome. So darker colors showing that the largest number of endangered mammal species are found. And it looks like in forest or rainforest, yellow savannah or grassland biome, there are like large number of endangered species are reported. So you can query that as well. You can also compare biodiversity across different biogeographic region. So this is also represented as an entity and it's color coded here. For example, yellow is neotropical realm, an orange paleoptic realm. And here is counting the number of animals reported, different animal species reported in each biogeographic region. And it looks like this yellow neotropical realm has rich biodiversity. And this is comparing the biomass across different species domain. Like plants, animals and it looks like plants dominate more than 3/4 of entire biomass on earth, while animals is a really small fraction. But surprisingly bacteria also kind of occupies a large part as well. We have also a host information for more than 190,000 species. So you can explore for example organisms that infect various crop species. And you get about like 5,000 different species that infect crops. And and this bar chart is illustrating comparing the phyla that organisms belong that infect crops. And it looks like a large number of fungus species are reported, but also includes some DNA viruses, bacteria, some roundworm species, flatworm species also reported here. And of course not just plants, but you can also query for human diseases that are caused by virus or any other microorganisms as well. So here showing the virus and associated diseases, for example like virus cause, COVID 19, smallpox, common flu, those are reported here. Another thing we can explore is food habits of species. We have predator information for over 7,000 species that you can query animals by predators. So for example, this one is specifically showing animals that are observed in Illinois that the Cooper's hog prey on. So you can see the different species of birds. Some mammals like squirrels, rats or even some reptiles are illustrated here in this image collage. And another nice thing is, you know, it's probably useful for the class biology class for kids who is learning food chains. You can easily visualize the food can. This example is just selectively shown animals observed in Grand Canyon National Park. And based on predator prey relationships you can easily create network graph network like this here. Like you can see the coyote on top of the hierarchy and including bobcat squirrels, birds, some beetles shown here. There are other astrological aspects of animals that we can explore like daily activity patterns like social behaviors, seasonal movement strategy and combination communication perception. So you can easily characterize animals by these some behavioral characteristics. So this is just one example. We can explore phylogenetic distribution of popular just animals and their clustered behavioral characteristics patterns. Red is showing like social behaviors, orange is locomotive behaviors, yellow is seasonal movement strategies and blue is some activity patterns. And you can see that some phylogenetically like similar animals share some like yeah, behavioral patterns. And we can easily look up animals by some characteristics. For example, you can find snake that have cryptic ability for their interpreted adaptation strategies. So these species of snakes have camouflage ability. Or here it's showing fish species that can perceive magnetic signals. And beyond behavioral characteristics, we can also look up animals by some reproductive features like mating system like bars types, sexual dimorphism or even different developmental strategies. I think this is cool example that some reptiles, in some reptiles their sex is determined based on temperature and those reptiles are illustrated here. And it's also not worth it that this type of query entity class syntax using entity class syntax is actually queried using this control echo and just type in your like natural language input. So for example, if you ask like some popular comparative queries like what are the 10 heaviest flight capable bars and this is easily converted to this entity class syntax and that in return can be plugged into this entity value core and give you a nice data set illustrating the species with their weight and same thing this is just asking what are the 50 longest living Vatable rates. And that's also converted as an implicitly defined entity class. And you can create like this plot like this as well. And beyond animals we also support the lookup of plants. So if you're a gardener or anybody interested in like finding plants to you grow in your garden, there's various properties to describe Gross condition of plants like vegetables and flowers. So for example some requirements rotoruants how easy to care if it's edible like companion plants pollinators. That kind of information is also affordable. Also some like quantitative values like how many days to harvest and how tall it can be. So for example this is finding plants based on growth conditions. I specifically select biennial vegetables that can grow in zone six USDA hardiness zone and of course soil texture. And you get a list of vegetables. And this USDA plant hardness zone is also described as an entity. So you get a nice data set as well some information about the zone. And if you are not sure about your hardness zone, you can also look up their zone by zip code. So for example zip code 61820 belongs to zone 5, specifically zone 5B. And we also have a nice graphical images to show where the zone 6 for example A can be found. And yeah, also we are constantly trying to expand this data set more information beyond animals and plants. And we recently introduced new entity types to just characterize fungus. So for example fungus cap type illustrates the physical characteristics of mushrooms so how they look like or like fungus edibility to characterize the edibility of mushroom. So for example you can specifically ask what are the like psychoactive mushroom that with spore print color. So just a quick summary. We have a new species data collections that encompass various biological, geographical and ecological knowledge of various organisms. And this hierarchical entity type syntax. And improved data lookup support in depth biological data exploration in the Wolfram language. And we are trying to expand biology knowledge base in many ways to like deliver something interesting for students and others to explore. So yeah, I finally like to thank my or my biology team and that's all for my talk. But I'd be happy to answer any questions or if you have any suggestions comments, please feel free to. Yeah, leave that comment in the chat. Thank you. So I have a question from Robot. Do you have those climate zones for different years? I would be interested to see how they are changing over time. Yeah, that's a really interesting question. Yeah, I also love to see that too. But unfortunately yeah, currently we don't have that kind of information. But I think yeah, that's an interesting solution. So it looks like almost time is up, but yeah, please feel free to leave any comments suggestions in the chat. And yeah, thank you so much for joining this session. Thank you.