New Biology Content in the Wolfram Language

Hello, My name is Keiko Irema, I'm a Wolfram Alpha developer in Scientific Content Group. I'm introducing new biology content accessible in the Wolfram language. Hi.

Hello.

So we are introducing new biology content accessible in the Wolfram language. So in this session we'd like to highlight three new features in biology area. I would like to introduce new species data including updated taxonomy information as well as support to query and explore various characteristics of various organisms. And we are lucky to have two biology experts in our team. Lina will be introducing dinosaur and prehistoric animal data sets and Cover loss will be introducing updated genomics data. And for species data we made update recently to include updated information. So new species data is going to be available very very soon, which is based on NCBI taxonomy database and providing hierarchical relational information for close to 2 million taxas so including almost 1 million animals. And great things about this new data set is that you can get more detailed taxonomic rank information. So in addition to traditional primary taxonomic sequence information for individual species, you can also get secondary rank level information so you can get more detailed like subphylums super order those kind of information is available so that allow us to get a better picture of a tree of life. So and also you can get a list of species belong to given taxonomic branch in 3 very easily. So for instance you can get the list of all species belong to this elephant family in taxonomic tree. And we also added support to query or group organisms by their specific properties like ecological, geological or other biological characteristics. So for instance you can get a group of species that is considered as fish and also that can perceive magnetic signal and you get a nice list of species with images. And we also added support to query various paraphyletic species group. So so those are group of organisms that belong to multiple branches in taxonomic tree. So example is otters. There is no single taxonomy to describe authors in current taxonomy database, so those are rather defined in paretic group. So you can actually query all species that are considered as otter in this way and you can query specific properties about and aggregate them. And so for instance you can access taxonomy graph property to actually confirm that these otter species belong to different branches in taxonomic 3. That's illustrating here and there are various properties you can access to analyze species. So for example conservation status information is available for over 70,000 species including the latest extinct endangered species information. So you can do some interesting thing like for example you can retrieve all terrestrial mammals and you can retrieve data set on conservation status and specific habitat, terrestrial biome or habitat they belong and use Classifier function to associate biomes with the likelihood of specific species having threatened sodas. So after also mentioned that we have a nice polygon information for each ecoregion or biome of species. So you can get illustrate a nice map in the dark area I think representing rainforest or forest biome there are more species in threatening conservation status. We also have like host and disease information of species. So for instance you can get all pathogens for various pathogens that can affect various animals or even plants. Or you can associate virus with common human diseases. And you can also illustrate like a network of host pathogen illustration like this. So we have a really nice interesting properties that you can explore ecological characteristics of animals. And there are also biological properties that you can explore. You can access nucleotide sequence length information or chromosome information from ncbi. For example here this is illustrating the species with the largest number of chromosomes. Surprisingly you can see pherons has the largest number of chromosomes. And if you are interested in analyzing genome sequence, you can actually retrieve reference sequence get genome accession number from ncbi. So for over a million of species you can get this external object that actually links to NCBI website and you can retrieve genome sequencing fast format so that you can do more sequence analysis. With this. You can explore geographic properties of species as well. You can look up animals by habitat or look up animals by observation locations. We have over 1.3 million data points that you can explore that are actually aligned to geo entities. Not only just countries, but ocean, mountain park, volcano and so on. For instance you can find the mammals specifically you can find in bugs and you get the list. Or reversely you can find the oceans where you can find dugongs for example and you can map it using geographics function for animals that can migrate. You can also analyze observe the migratory patterns whenever. Here we have monarchy butterfly and you can retrieve the geopoints where you can find these butterflies by date so that you can actually analyze the migratory pattern of these butterflies. And lastly we also have properties to explore food habits of animals. So you can access predator information of animals found in Illinois for example here and you can get a list of animals that Cooper's hawk prey on and you get nice list. Or using the same predator properties you can create the food network of animals found in Grand Canyon National Park. So you can do really cool things with this new data set. So just a quick summary of the species data. We have a really nice new dataset that allows to do more complex lookups and you can explore various characteristics. There are even more that I couldn't mention in this talk, but there's really interesting things you can do with this new data set. So that's all I have for species part. And now I'd like to hand this to Lena for the next part.

Hi, I'm Lena and I've been working on dinosaur and prehistoric animals data sets, the odatabases. This is the outline of this short presentation. First a short description of the database sources, then some important things about the new diagnosis and prehistoric animals databases. Some graph analysis we can do with this new information and finally a summary. So the first thing to say about the sources is that they are open source. We use the compact Tessellosaurus database for dinosaur taxa. Also use the paleobiology database for both dinosaur and prehistoric animals databases or taxa. And we also got some information from Wikidata. Okay, first for the dinosaur new databases we have new taxa and new values for the previous properties. So for example, in the already existing dinosaur database in Wolfram Alpha there is around 1,500 taxa and we added with the new database around 1,700 new taxon. Also for the already existing properties for these dinosaur entities we have new values. So for example, for the discovery country, the country for which the fossil are reported already existing database we covered these countries. But with the new database we covered more countries for both the already existing taxa and for the new taxa. Another example is that we for geological period property, for example, we increase with the new database the number of taxa that is reported for each one of these geological periods. So in orange is the tax the number of taxa that is reported for each of the geological periods in the already existing data Wolfram Alpha database. And in purple is the new number of taxa reported for each one of these geological periods. Also for these new dinosaur data ways we have new properties with these new properties. These new properties give us a higher resolution of the location in which the fossils are found, such as coordinates, states and city as you can see in this plot. Also we have more information of the geological time to which the new taxable loans the new and the already existing taxa ballons to e.g. geological formations. For example, here is the 10 most frequent of the 10 most popular formation of the dinosaur database and also a new property that is the geological age to which the taxa belongs to that it was already processed at entity by Jeff. You can find their properties. Then for prehistoric animals database this is a completely new database. In Wolfpren Alpha we have around 34,000 taxa including species, genus and so on. We have new properties additional to the already mentioned for dinosaur database properties such as reproduction and this is his balance for reproduction property. Also new property as behavior. So for each one of the new tax and prehistoric animals. Finally, we can also make some graph analysis. You know, this new information. For example, here I'm showing the timeline plot of the geological periods of the 10 most popular dinosaur. For example, for Stegosaurus, these new databases reported that they belong to middle and late Jurassic and also to the early and late Cretaceous. As you can see here in the timeline plot. Another interesting graph analysis we can make is for example this here this clustering of the 10 most popular prehistoric animals. And here this Dendogram cluster the taxa which is species level according with their similarities in for example diet, environment, behavior and reproduction. The black box represent the particular barrier for each one of these properties in colors to which each one of these taxa belongs to. We also can make some. We can find some communities of a graph, for example, this graph. In this graph, the vertex are the families from dinosaur database in colors. Each of the families are represented in colors and the numbers is the geological period to which some of the taxa belongs to. For example, this family here in green have some taxa that belongs to the phi geological period, which is late Triassic epoch. Also, some of their taxa belongs to the seven period, which is also Triassic. About the middle. In this graph, for example, we can see that this community show a variety of families, dinosaur families that belongs to that. Sorry. With families that quit taxa that belongs to the early Cretaceous. For example, a similar graph. This is a similar graph to the previous one, but here instead of geological period, I put a species behavior. So here we find a community in which all these families, dinosaur families have taxa that is that have that their taxa are social, for example, with this kind of behavior. Also, for example, this family in green have taxa that is for example flight capable, or some of them are social, or some of them are solitary. For example. Finally, a summary of the presentation with this new we curated so the dinosaur and prehistoric animal databases that have new taxa, new properties, new values for the previous properties. Also we curated, although I didn't show so much about them, but also created some geological formation databases with all this information. With all this new information, we can make a variety of graph analysis or the analysis that you want to make. Thanks.

So I'll talk to you about the updates to genomics data. So in summary, updates to genomics data include access to gene cohomology information and associating human gene properties to specific assemblies, as well as providing genome assembly statistics. So first I'll talk To you about the updates to gene homology information. Homologs of a gene entity can be supplied as a list of gene entities. For example, if we want to see the genes homologous the homologous genes to the tumor protein p53 gene in humans, we can do this and we get a list of genes that are in chimpanzee, cow, mouse, rat and zebrafish. Note that we currently only have gene entities for a group of model organisms. With this, we can do a lot of things. One thing we can do is we can access the reference sequences for these genes. I'm putting them in a bio sequence so it's easier to visualize. With this, we can do things like compute the distances between sequences or the similarities between sequences, and create a dendrogram to show how genes from closely related species tend to be closer to each other. Here we can see that the chimp and human are in the same group as well as the mouse and rats, which reflect the evolutionary relationships we know exist. Another way to see this is also using the clustering tree. It gives us a similar graph. The difference here is that the lengths of the edges do not reflect the distances, but this is another way we can visualize the state. So one of the other updates that are coming is that gene properties are now tied to the latest reference genome and unless specified, a query will return data relative to the latest assembly induced by Wolfram alpha. So, for example, here if we want to get the reference sequence for the tumor protein P50C gene, we can ask for it and we can also specify this BILD38P13, which is also the latest assembly, and we can see that they are the same sequence data relative to previously supported assemblies can still be accessed by using the genome assembly as a qualified Note that we only support reference sequence information for the previous assembly 37P3 and 36.3. And here is how we can obtain it using genome assembly and specifying disassembly. With this, we can do different things like we can see how the reference sequences have changed along with new gene ROM assemblies. So in this case, I'm comparing the latest reference sequence with the one from 37p3 for this p53 gene. And when we align them, we see that there are two sections which are different from each other, one which has lost this short 55 base sequence and this other 19 base sequence. And along with these updates, we can also retrieve various statistics for the assemblies we support. So if we use the entity value data set, we can get a data set with all these properties where we can see some descriptive properties like the names and associated species, some properties related to the completeness or quality of the assembly and the accession numbers, and a few other things. With this we can do different things. One of them is measuring how assemblies have changed over time. So a measure of the quality of an assembly is a contig N50 value, where generally speaking a larger N50 value indicates higher quality between assemblies of similar length. So here we get the contig N50 values by release date and if we plot them we can see how they have increased over time from around a value of 2-point-something million to around 5-point-something million. Another measure can be the percentage of sequence composed by gaps and we can also see how it has changed over time. So doing something very similar we can see how the percentage of gaps has gone from around 7.6% to less than 5% currently. So in summary, the updates to genomics data include that gene homology information is now accessible in or will be accessible in the Wolfram language using the Entity framework. Also, gene properties are now tied to the latest reference genome and data relative to previously supported assemblies are still available and genome assembly statistics such as release date and percentage of gaps are provided.