An accessible, timely method for identifying medications for repurposing: hypothesis generating; testing and validation

Maximin Lange; Eoin Gogarty; Meredith Martyn; Philip Braude; Feras Fayez; Nikolaos Koutsouleris; Oliver Howes; Ricardo Twumasi; Ben Carter

doi:10.1186/s44247-025-00170-8

An accessible, timely method for identifying medications for repurposing: hypothesis generating; testing and validation

Maximin Lange^*, Eoin Gogarty, Meredith Martyn, Philip Braude, Feras Fayez, Nikolaos Koutsouleris, Oliver Howes, Ricardo Twumasi, Ben Carter

^*Corresponding author for this work

North Bristol NHS Trust

Research output: Contribution to journal › Article › peer-review

25 Downloads (Pure)

Abstract

Background
Drug development is lengthy and costly, making drug repurposing an attractive alternative. Identifying repurposing candidates from vast biomedical literature is challenging. Natural language processing (NLP) offers potential for literature-based discovery. We present and evaluate a novel, accessible NLP-based method using the Word2Vec algorithm to identify, test, and validate candidate medications for repurposing, demonstrated by seeking treatments for psychotic disorders.

Results
A Word2Vec model trained on 2.3 million PubMed abstracts (2000–2023) identified potential repurposing candidates based on cosine similarity to a known antipsychotic drug. We tested one candidate, a cephalosporin antibiotic, across independent datasets: MIMIC-IV, CRIS, and BRATECA. Cephalosporin antibiotics in MIMIC-IV demonstrated a reduced hazard ratio (aHR) for psychosis hospitalisation overall (0.94, 95% CI: 0.90–0.99) and more substantially for severe mental illness (0.52, 95% CI: 0.45–0.60). However, CRIS showed an increased risk (HR: 3.56, 95% CI: 2.66–4.77). BRATECA lacked suitable diagnostic data for analysis.

Conclusions
This methodological framework demonstrates the potential for machine learning approaches to systematically identify drug repurposing candidates, while highlighting population-specific variations in therapeutic effectiveness that warrant caution in translational applications. Our findings should not be interpreted as cephalosporins to be recommended for or against treating psychosis. Our findings merely offer validity to the application of our repurposing methodology, and should serve as a foundation for further investigation rather than direct clinical application.

Original language	English
Article number	30
Journal	BMC Digital Health
Volume	3
DOIs	https://doi.org/10.1186/s44247-025-00170-8
Publication status	Published - 22 May 2025

Keywords

drug repurposing
natural language processing
Word2Vec
machine learning
psychosis
schizophrenia
cephalosporins
electronic health records
MIMIC-IV
Crise financière

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1186/s44247-025-00170-8Licence: CC BY-NC-SA

Author copy Lange et al 2025 BMCAccepted author manuscript, 353 KBLicence: CC BY-NC-SA
Lange et al 2025 BMCAccepted author manuscript, 1.81 MBLicence: CC BY-NC-SA

Cite this

@article{4efa1e54d6e74f7498c6b51fafcab257,

title = "An accessible, timely method for identifying medications for repurposing: hypothesis generating; testing and validation",

abstract = "BackgroundDrug development is lengthy and costly, making drug repurposing an attractive alternative. Identifying repurposing candidates from vast biomedical literature is challenging. Natural language processing (NLP) offers potential for literature-based discovery. We present and evaluate a novel, accessible NLP-based method using the Word2Vec algorithm to identify, test, and validate candidate medications for repurposing, demonstrated by seeking treatments for psychotic disorders.ResultsA Word2Vec model trained on 2.3 million PubMed abstracts (2000–2023) identified potential repurposing candidates based on cosine similarity to a known antipsychotic drug. We tested one candidate, a cephalosporin antibiotic, across independent datasets: MIMIC-IV, CRIS, and BRATECA. Cephalosporin antibiotics in MIMIC-IV demonstrated a reduced hazard ratio (aHR) for psychosis hospitalisation overall (0.94, 95% CI: 0.90–0.99) and more substantially for severe mental illness (0.52, 95% CI: 0.45–0.60). However, CRIS showed an increased risk (HR: 3.56, 95% CI: 2.66–4.77). BRATECA lacked suitable diagnostic data for analysis.ConclusionsThis methodological framework demonstrates the potential for machine learning approaches to systematically identify drug repurposing candidates, while highlighting population-specific variations in therapeutic effectiveness that warrant caution in translational applications. Our findings should not be interpreted as cephalosporins to be recommended for or against treating psychosis. Our findings merely offer validity to the application of our repurposing methodology, and should serve as a foundation for further investigation rather than direct clinical application.",

keywords = "drug repurposing, natural language processing, Word2Vec, machine learning, psychosis, schizophrenia, cephalosporins, electronic health records, MIMIC-IV, Crise financi{\`e}re",

author = "Maximin Lange and Eoin Gogarty and Meredith Martyn and Philip Braude and Feras Fayez and Nikolaos Koutsouleris and Oliver Howes and Ricardo Twumasi and Ben Carter",

year = "2025",

month = may,

day = "22",

doi = "10.1186/s44247-025-00170-8",

language = "English",

volume = "3",

journal = "BMC Digital Health",

issn = "2731-684X",

publisher = "Springer Nature",

}

TY - JOUR

T1 - An accessible, timely method for identifying medications for repurposing: hypothesis generating; testing and validation

AU - Lange, Maximin

AU - Gogarty, Eoin

AU - Martyn, Meredith

AU - Braude, Philip

AU - Fayez, Feras

AU - Koutsouleris, Nikolaos

AU - Howes, Oliver

AU - Twumasi, Ricardo

AU - Carter, Ben

PY - 2025/5/22

Y1 - 2025/5/22

N2 - BackgroundDrug development is lengthy and costly, making drug repurposing an attractive alternative. Identifying repurposing candidates from vast biomedical literature is challenging. Natural language processing (NLP) offers potential for literature-based discovery. We present and evaluate a novel, accessible NLP-based method using the Word2Vec algorithm to identify, test, and validate candidate medications for repurposing, demonstrated by seeking treatments for psychotic disorders.ResultsA Word2Vec model trained on 2.3 million PubMed abstracts (2000–2023) identified potential repurposing candidates based on cosine similarity to a known antipsychotic drug. We tested one candidate, a cephalosporin antibiotic, across independent datasets: MIMIC-IV, CRIS, and BRATECA. Cephalosporin antibiotics in MIMIC-IV demonstrated a reduced hazard ratio (aHR) for psychosis hospitalisation overall (0.94, 95% CI: 0.90–0.99) and more substantially for severe mental illness (0.52, 95% CI: 0.45–0.60). However, CRIS showed an increased risk (HR: 3.56, 95% CI: 2.66–4.77). BRATECA lacked suitable diagnostic data for analysis.ConclusionsThis methodological framework demonstrates the potential for machine learning approaches to systematically identify drug repurposing candidates, while highlighting population-specific variations in therapeutic effectiveness that warrant caution in translational applications. Our findings should not be interpreted as cephalosporins to be recommended for or against treating psychosis. Our findings merely offer validity to the application of our repurposing methodology, and should serve as a foundation for further investigation rather than direct clinical application.

AB - BackgroundDrug development is lengthy and costly, making drug repurposing an attractive alternative. Identifying repurposing candidates from vast biomedical literature is challenging. Natural language processing (NLP) offers potential for literature-based discovery. We present and evaluate a novel, accessible NLP-based method using the Word2Vec algorithm to identify, test, and validate candidate medications for repurposing, demonstrated by seeking treatments for psychotic disorders.ResultsA Word2Vec model trained on 2.3 million PubMed abstracts (2000–2023) identified potential repurposing candidates based on cosine similarity to a known antipsychotic drug. We tested one candidate, a cephalosporin antibiotic, across independent datasets: MIMIC-IV, CRIS, and BRATECA. Cephalosporin antibiotics in MIMIC-IV demonstrated a reduced hazard ratio (aHR) for psychosis hospitalisation overall (0.94, 95% CI: 0.90–0.99) and more substantially for severe mental illness (0.52, 95% CI: 0.45–0.60). However, CRIS showed an increased risk (HR: 3.56, 95% CI: 2.66–4.77). BRATECA lacked suitable diagnostic data for analysis.ConclusionsThis methodological framework demonstrates the potential for machine learning approaches to systematically identify drug repurposing candidates, while highlighting population-specific variations in therapeutic effectiveness that warrant caution in translational applications. Our findings should not be interpreted as cephalosporins to be recommended for or against treating psychosis. Our findings merely offer validity to the application of our repurposing methodology, and should serve as a foundation for further investigation rather than direct clinical application.

KW - drug repurposing

KW - natural language processing

KW - Word2Vec

KW - machine learning

KW - psychosis

KW - schizophrenia

KW - cephalosporins

KW - electronic health records

KW - MIMIC-IV

KW - Crise financière

U2 - 10.1186/s44247-025-00170-8

DO - 10.1186/s44247-025-00170-8

M3 - Article

SN - 2731-684X

VL - 3

JO - BMC Digital Health

JF - BMC Digital Health

M1 - 30

ER -

An accessible, timely method for identifying medications for repurposing: hypothesis generating; testing and validation

Abstract

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this