Automated recognition of functional compound-protein relationships in literature

Döring, Kersten; Qaseem, Ammar; Becer, Michael; Li, Jianyu; Mishra, Pankaj; Gao, Mingjie; Kirchner, Pascal A.; Sauter, Florian; Telukunta, Kiran K.; Moumbock, Aurélien F. A.; Thomas, Philippe; Günther, Stefan

doi:10.1371/journal.pone.0220925

Cited by 3 publications

(2 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To obtain the MLEs of maF * 1 and maF * 2 , we first substitute p a.a , p .aa , p a.. , p .a. and p ..a by their MLE's to get MLE's of maP and maR and use these in (9) and (10):…”

Section: Test Statistic For Comparing Two Maf * Smentioning

confidence: 99%

“…Although

{F}_1

‐scores for binary and multi‐class classifications have been originally used for measuring the performance of text classification in the field of information retrieval or of a classifier in machine learning, it has become frequently used in medicine 7‐14 . Some statistical methods for inference have been proposed for the binary

{F}_1

‐score, 15 and the methods for estimating confidence intervals of the micro‐averaged

{F}_1

‐scores and macro‐averaged

{F}_1

‐scores has been developed 16,17 .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Hypothesis testing procedure for binary and multi‐class F₁‐scores in the paired design

Takahashi,

Yamamoto,

Kuchiba

et al. 2023

Statistics in Medicine

View full text Add to dashboard Cite

In modern medicine, medical tests are used for various purposes including diagnosis, disease screening, prognosis, and risk prediction. To quantify the performance of the binary medical test, we often use sensitivity, specificity, and negative and positive predictive values as measures. Additionally, the ‐score, which is defined as the harmonic mean of precision (positive predictive value) and recall (sensitivity), has come to be used in the medical field due to its favorable characteristics. The ‐score has been extended for multi‐class classification, and two types of ‐scores have been proposed for multi‐class classification: a micro‐averaged ‐score and a macro‐averaged ‐score. The micro‐averaged ‐score pools per‐sample classifications across classes and then calculates the overall ‐score, whereas the macro‐averaged ‐score computes an arithmetic mean of the ‐scores for each class. Additionally, Sokolova and Lapalme gave an alternative definition of the macro‐averaged ‐score as the harmonic mean of the arithmetic means of the precision and recall over classes. Although some statistical methods of inference for binary and multi‐class ‐scores have been proposed, the methodology development of hypothesis testing procedure for them has not been fully progressing yet. Therefore, we aim to develop hypothesis testing procedure for comparing two ‐scores in paired study design based on the large sample multivariate central limit theorem.

show abstract

“…To obtain the MLEs of maF * 1 and maF * 2 , we first substitute p a.a , p .aa , p a.. , p .a. and p ..a by their MLE's to get MLE's of maP and maR and use these in (9) and (10):…”

Section: Test Statistic For Comparing Two Maf * Smentioning

confidence: 99%

“…Although

{F}_1

{F}_1

‐score, 15 and the methods for estimating confidence intervals of the micro‐averaged

{F}_1

‐scores and macro‐averaged

{F}_1

‐scores has been developed 16,17 .…”

Section: Introductionmentioning

confidence: 99%

Hypothesis testing procedure for binary and multi‐class F₁‐scores in the paired design

Takahashi,

Yamamoto,

Kuchiba

et al. 2023

Statistics in Medicine

View full text Add to dashboard Cite

show abstract

CPRiL: compound–protein relationships in literature

Qaseem

Günther

2022

Bioinformatics

View full text Add to dashboard Cite

Newly discovered functional relationships of (bio-)molecules are a key component in molecular biology and life science research. Especially in the drug discovery field, knowledge of how small molecules associate with proteins plays a fundamental role in understanding how drugs or metabolites can affect cells, tissues, and human metabolism. Finding relevant information about these relationships among the huge number of published articles is becoming increasingly challenging and time-consuming. On average, more than 25,000 new (bio-)medical articles are added to the literature database PubMed weekly. In this work, we present a new web server (CPRiL) which provides information on functional relationships between small molecules and proteins in literature. Currently, CPRiL contains ∼465,000 unique names and synonyms of small molecules, ∼100,000 unique proteins, and more than 9 million described functional relationships between these entities. The applied BioBERT machine learning model for the determination of functional relationships between small molecules and proteins in texts was extensively trained and tested. On a related benchmark, CPRiL yielded a high performance, with an F1-score of 84.3%, precision of 82.9%, and recall of 85.7%. Availability CPRiL is freely available at https://www.pharmbioinf.uni-freiburg.de/cpril.

show abstract

Discovering drug–target interaction knowledge from biomedical literature

Hou

Xia

et al. 2022

Bioinformatics

View full text Add to dashboard Cite

Motivation The Interaction between Drugs and Targets (DTI) in human body plays a crucial role in biomedical science and applications. As millions of papers come out every year in the biomedical domain, automatically discovering DTI knowledge from biomedical literature, which are usually triplets about drugs, targets and their interaction, becomes an urgent demand in the industry. Existing methods of discovering biological knowledge are mainly extractive approaches that often require detailed annotations (e.g., all mentions of biological entities, relations between every two entity mentions, etc.). However, it is difficult and costly to obtain sufficient annotations due to the requirement of expert knowledge from biomedical domains. Results To overcome these difficulties, we explore an end-to-end solution for this task by using generative approaches. We regard the DTI triplets as a sequence and use a Transformer-based model to directly generate them without using the detailed annotations of entities and relations. Further, we propose a semi-supervised method, which leverages the aforementioned end-to-end model to filter unlabeled literature and label them. Experimental results show that our method significantly outperforms extractive baselines on DTI discovery. We also create a dataset, KD-DTI, to advance this task and release it to the community. Availability Our code and data are available at https://github.com/bert-nmt/BERT-DTI. Supplementary information Supplementary data is available at Bioinformatics. online.

show abstract

Automated recognition of functional compound-protein relationships in literature

Cited by 3 publications

References 36 publications

Hypothesis testing procedure for binary and multi‐class F₁‐scores in the paired design

Hypothesis testing procedure for binary and multi‐class F₁‐scores in the paired design

CPRiL: compound–protein relationships in literature

Discovering drug–target interaction knowledge from biomedical literature

Contact Info

Product

Resources

About

Automated recognition of functional compound-protein relationships in literature

Cited by 3 publications

References 36 publications

Hypothesis testing procedure for binary and multi‐class F1‐scores in the paired design

Hypothesis testing procedure for binary and multi‐class F1‐scores in the paired design

CPRiL: compound–protein relationships in literature

Discovering drug–target interaction knowledge from biomedical literature

Contact Info

Product

Resources

About

Hypothesis testing procedure for binary and multi‐class F₁‐scores in the paired design

Hypothesis testing procedure for binary and multi‐class F₁‐scores in the paired design