Data-Driven Learned Metric Index: An Unsupervised Approach

Slanináková, Terézia; Antol, Matej; OIha, Jaroslav; Kana, Vojtech; Dohnal, Vlastislav

doi:10.1007/978-3-030-89657-7_7

Cited by 6 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This research direction was later expanded to the domain of complex data where data items are compared using a distance function. The distance expresses complex similarity beyond mere comparison of two integers, as shown in (9, 12). The latter two methods in conjunction with (10) establish the basis of the indexing solution presented in here.…”

Section: Description Of the Web Servermentioning

confidence: 99%

AlphaFind: Discover structure similarity across the entire known proteome

Prochazka,

Slaninakova,

Olha

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

AlphaFind is a web-based search engine that provides fast structure-based retrieval in the entire set of AlphaFold DB structures. Unlike other protein processing tools, AlphaFind is focused entirely on tertiary structure, automatically extracting the main 3D features of each protein chain and using a machine learning model to find the most similar structures. This indexing approach and the 3D feature extraction method used by AlphaFind have both demonstrated remarkable scalability to large datasets as well as to large protein structures. The web application itself has been designed with a focus on clarity and ease of use. The searcher accepts any valid Uniprot ID, PDB ID or gene symbol as input, and returns a set of similar protein chains from AlphaFold DB, including various similarity metrics between the query and each of the retrieved results. In addition to the main search functionality, the application provides 3D visualizations of protein structure superpositions in order to allow researchers to instantly analyze the structural similarity of the retrieved results. The AlphaFind web application is available online for free and without any registration at https://alphafind.fi.muni.cz.

show abstract

Section: Description Of the Web Servermentioning

confidence: 99%

AlphaFind: Discover structure similarity across the entire known proteome

Prochazka,

Slaninakova,

Olha

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Following the architectural design of RMI, we proposed the Learned metric index (LMI) [1], which can use a series of arbitrary machine learning models to solve the classification problem by learning a pre-defined partitioning scheme. This was later extended to a fully unsupervised (data-driven) version introduced in [18], which is utilized in this work.…”

Section: Related Workmentioning

confidence: 99%

“…We explored different architectural setups -both in terms of the number of nodes at each level (index breadth), as well as the number of levels (index depth). As the learned models, we explored K-Means, Gaussian Mixture Models, and K-Means in combination with Logistic regression (see [18] for details regarding the model setups). For the sake of compactness, in the experimental evaluation we only present the results achieved with the best-performing setup -a two-level LMI structure with arity of 256 on level 1 and 64 on level 2 (i.e., 256 root descendants, each of them with 64 child nodes), with K-Means chosen as the partitioning algorithm.…”

mentioning

confidence: 99%

Learned Indexing in Proteins: Substituting Complex Distance Calculations with Embedding and Clustering Techniques

Oľha¹,

Slanináková²,

Martin³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Despite the constant evolution of similarity searching research, it continues to face the same challenges stemming from the complexity of the data, such as the curse of dimensionality and computationally expensive distance functions. Various machine learning techniques have proven capable of replacing elaborate mathematical models with combinations of simple linear functions, often gaining speed and simplicity at the cost of formal guarantees of accuracy and correctness of querying. The authors explore the potential of this research trend by presenting a lightweight solution for the complex problem of 3D protein structure search. The solution consists of three steps -(i) transformation of 3D protein structural information into very compact vectors, (ii) use of a probabilistic model to group these vectors and respond to queries by returning a given number of similar objects, and (iii) a final filtering step which applies basic vector distance functions to refine the result.

show abstract