LocText: relation extraction of protein localizations to assist database curation

Cejuela, Juan Miguel; Vinchurkar, Shrikant; Goldberg, Tatyana; Shankar, Madhukar Sollepura Prabhu; Baghudana, Ashish; Bojchevski, Aleksandar; Uhlig, Carsten; Ofner, André; Raharja-Liu, Pandu; Jensen, Lars Juhl; Rost, Burkhard

doi:10.1186/s12859-018-2021-9

Cited by 129 publications

(26 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We hypothesize that the evidence for annotations is provided in the k-nearest sentences to the sentence where the protein (or its coding gene) is mentioned. Indeed, as shown by Cejuela et al [14], the k-1 sentences accounts for 89% of all unique relationships in the case of protein location evidence. To identify the candidate evidence sentences, occurrence of protein features, such as accession identifier, protein name (recommended, alternative and short), gene name and their synonyms, are searched in the sentences.…”

Section: Candidate Sentences For Annotation Evidencementioning

confidence: 88%

“…Text mining has also supported more specific curation tasks, such as protein localization. LocText, for example, implements a NER and RE for proteins based on SVM, achieving 86% precision (56% F1-score) [14]. To address the common issue of class imbalance in biocuration, an ensemble of SVM classifiers along with random under-sampling were proposed for automatically identifying relevant papers for curation in the Gene Expression Database [15].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

UPCLASS: a Deep Learning-based Classifier for UniProtKB Entry Publications

Teodoro

Knafou

Naderi

et al. 2019

Preprint

View full text Add to dashboard Cite

AbstractIn the UniProt Knowledgebase (UniProtKB), publications providing evidence for a specific protein annotation entry are organized across different categories, such as function, interaction and expression, based on the type of data they contain. To provide a systematic way of categorizing computationally mapped bibliography in UniProt, we investigate a Convolution Neural Network (CNN) model to classify publications with accession annotations according to UniProtKB categories. The main challenge to categorize publications at the accession annotation level is that the same publication can be annotated with multiple proteins, and thus be associated to different category sets according to the evidence provided for the protein. We propose a model that divides the document into parts containing and not containing evidence for the protein annotation. Then, we use these parts to create different feature sets for each accession and feed them to separate layers of the network. The CNN model achieved a F1-score of 0.72, outperforming baseline models based on logistic regression and support vector machine by up to 22 and 18 percentage points, respectively. We believe that such approach could be used to systematically categorize the computationally mapped bibliography in UniProtKB, which represents a significant set of the publications, and help curators to decide whether a publication is relevant for further curation for a protein accession.

show abstract

Section: Candidate Sentences For Annotation Evidencementioning

confidence: 88%

Section: Introductionmentioning

confidence: 99%

UPCLASS: a Deep Learning-based Classifier for UniProtKB Entry Publications

Teodoro

Knafou

Naderi

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Overall, most prediction methods came relatively close ( Fig. 2A: the more white the fewer the mistakes) to estimating the 7-class location spectrum for the new data sets that had not been used for developing the methods (HPA: new data from The Human Protein Atlas, and the LocText text mining results [9]). When error-correcting the predicted spectra according to the performance confusion matrix [14], the observations and predictions became more similar for most methods (Fig.…”

Section: Accurate Predictions Of Location Spectrum For Organismsmentioning

confidence: 88%

“…This resulted in a set of 5,563 proteins with experimental annotations from Swiss-Prot (release 2017_1; human proteome release up000005640), and in 12,036 from The Human Protein Atlas (version 15; confined to 32 localization classes). We had access to one additional set of experimental annotation in GO format extracted from scientific literature by the tool LocText [9].…”

Section: Experimental Annotationsmentioning

confidence: 99%

See 1 more Smart Citation

Spectrum of protein localization in proteomes captures evolutionary relation between species

Marot-Lassauzaie

Goldberg

Rost

2019

Preprint

Self Cite

View full text Add to dashboard Cite

The native subcellular localization or cellular compartment of a protein is the one in which it acts most often; it is one aspect of protein function. Do ten eukaryotic model organisms differ in their location spectrum, i.e. the fraction of its proteome in each of its seven major compartments? As experimental annotations of locations remain biased and incomplete, we need prediction methods to answer this question. To gauge the bias of prediction methods, we merged all available experimental annotations for the human proteome. In doing so, we found important values in both Swiss-Prot and the Human Protein Atlas (HPA). After systematic bias corrections, the complete but faulty prediction methods appeared to be more appropriate to compare location spectra between species than the incomplete more accurate experimental data. This work compared the location spectra for ten eukaryotes: Homo sapiens, Gorilla gorilla, Pan troglodytes, Mus musculus, Rattus norvegicus, Drosophila melanogaster, Anopheles gambiae, Caenorhabitis elegans, Saccharomyces cerevisiae and Schizosaccharomyces pombe. Overall, the predicted location spectra were similar. However, the detailed differences were significant enough to plot trees and 2D (PCA) maps relating the ten organisms using a simple Euclidean distance in seven states, corresponding to the seven studied localization classes. The relations based on the simple predicted location spectra captured aspects of cross-species comparisons usually revealed only by much more detailed evolutionary comparisons.

show abstract

High‐Throughput Automated Subcellular Localisation

Chalkley

Kelly

Mysior

et al. 2020

Encyclopedia of Life Sciences

View full text Add to dashboard Cite

Defining the subcellular localisation of the proteome for an organism of interest is a critical next step following genome sequencing. Knowledge of protein subcellular localisation provides insight into the functionality of the normal cell, as well during disease states. However, the presence of gene isoforms, alternative splicing and posttranslational modifications significantly increase the number of protein variants encoded by a single gene, making this a complex task. In the last 20 years, parallel approaches using fractionation and mass spectrometry, synthesis of large libraries of open reading frames fused to genes encoding fluorescent proteins, as well as production of thousands of antibodies have all contributed to the systematic analysis of protein localisation. Alongside these methods, improved bioinformatic predictors, machine learning and deep learning algorithms have also evolved as essential tools. A combinatorial approach of these methods now brings us close to systematically defining the subcellular proteome for many organisms. Key Concepts Subcellular localisation is a critical determinant in understanding protein function. Data from genome sequencing projects provide the fundamental information from which approaches to understand protein localisation can be initiated. Parallel approaches using fluorescence microscopy are being applied in a high‐throughput manner to systematically reveal the subcellular localisation of large numbers of proteins in different cells. The primary techniques to determine protein localisation are mass spectrometry‐based proteomics, production of antibodies and expression of fluorescently tagged proteins. There is increasing use of computational biology tools to aid the automated classification of subcellular localisation. Large image datasets can be interrogated by machine learning software algorithms to automatically classify proteins to specific localisations. Deep learning methods, which can work independently of training datasets, have become the newest tool to automatically assign protein localisation from image sets. Automated approaches combining both experimental and computational methods are likely to become the primary means by which subcellular localisation is determined from new cell systems.

show abstract

LocText: relation extraction of protein localizations to assist database curation

Cited by 129 publications

References 48 publications

UPCLASS: a Deep Learning-based Classifier for UniProtKB Entry Publications

UPCLASS: a Deep Learning-based Classifier for UniProtKB Entry Publications

Spectrum of protein localization in proteomes captures evolutionary relation between species

High‐Throughput Automated Subcellular Localisation

Contact Info

Product

Resources

About