|
|
A neural network for reconstruction of extinct animals
Pešek, David ; Bilík, Šimon (referee) ; Jirsík, Václav (advisor)
This work was focused on designing, learning and evaluating an artificial neural network for reconstructing extinct species. First, the main element of the proposed artificial neural network, i.e., the generative model, was selected. Given their excellent performance in the field of image generation, the class of diffusion models reasonably seemed to be the right choice. Specifically, the Stable diffusion model was chosen. One of the initial steps of the work was to create a training set for the proposed model. The animal images needed to be paired with some labels that could be used to identify the animal. For this purpose, the cytochrome c oxidase subunit I genes of the given animals were used. Furthermore, the sequential transformer model GPT-2, which is learned on the training set of human natural language, was used. This model was used to encode the DNA sequences into a vector form in which the semantics and context between the different parts of the DNA sequence were captured. The models would be very difficult to learn from scratch due to the large training set size required and the computational and time requirements. Thus, the GPT-2 model was only learned on the training set of DNA sequences of the passeriformes order, and the diffusion model itself was learned on pairs of images of these animals and DNA sequences encoded by the GPT-2 model. To generate the images, the original DNA sequences that resembled the sequences from the training set were generated using GPT-2. The encoding of these sequences was then passed to the diffusion model, which generated the images itself. The method of generating new DNA sequences using the GPT-2 model is based on the idea that the generated DNA sequence partially resembles the DNA sequences from the training set. Such experimentally generated DNA sequences may resemble DNA sequences of extinct ancestors or relatives of the passeriformes order. The model was in some cases able to generate images that could be considered as animal species , but it should be noted that often the generated images could not be considered as animal reconstructions. The success rate of generating a decent animal image was approximately 10%. The functionality of the model was also tested on a test set of DNA sequences of animals of several orders that fall under the class of birds as well as the order of passeriformes. The success rate of generating a reconstruction that could be compared to a photograph was around 5%.
|
|
|
The Current Extinction of species
Schätz, Filip ; Storch, David (advisor) ; Hulva, Pavel (referee)
This bachelor thesis deals with the current process of species extinction. The main aim of the thesis is to answer the question how we can accurately describe this process. The understanding of this process is based on the limited sample of extinct species known to us, provided by the IUCN Red List. The sheer number of extinct species is affected by the use of different species concepts, the synonymy of binomial names and the difficulty of tracking individual extinction events. The Red List sample is subject to strong selection bias, so currently there are only reliable records of extinctions of groups of birds and mammals. These extinctions are mainly associated with their occurrence on islands. For other taxa, records of extinctions are sporadic, so the status of the current extinction process in these groups is unknown. It is not possible in most cases to relate to these groups the knowledge resulting from the study of different taxa. Estimates that attempt to describe the process beyond our knowledge also suffer from shortcomings. While the problems of expressing actual extinction rates stem from our ignorance of biodiversity, the inaccuracies of the SAR curve back-extrapolation method are mainly due to the attempt to generalise the overly complex effect of range loss on species numbers. The...
|
|
|
The Black and Red lists of the Giant Mountains flora: Do the extinct and endangered plants have common properties?
Blahník, Jan ; Čihař, Martin (referee) ; Krahulec, František (advisor)
A list of extinct and endangered species of the Giant Mountains was published in 2009, which presents a uniform view of vascular plants on the Polish and the Czech sides of the mountains. This Bachelor's Thesis is focused on the most endangered and the extinct species, for which common properties were sought, while it was not clear, whether such properties actually existed. The Bachelor's Thesis evaluates factors that could be traced in literature. Data concerning the type of propagation, which is of key importance for the maintenance of species, however, are not stated in literature for the absolute majority of cases. Easily traceable data concerned the properties of the subsoil, biotope types and properties, life form (was not eventually evaluated), phytogeographic areas, degrees of vegetation and endemism. It cannot be expected that a comparison of the results would show conformity anywhere in the region of 100%, yet certain signs of common properties can be deduced from some results. For example, for one-fifth to one-third of C1 and A1 category species a common property is an acid substrate, 30-40% of C1 category species occur in oreophytic areas, for some 30% of extinct and endangered species, the common sign is their occurrence in meadow sites, 30-40% of critically endangered species occur in...
|
guest ::
