Advances in the development of PubCaseFinder, including the new application programming interface and matching algorithm

Fujiwara, Toyofumi; Shin, Jae-Moon; Yamaguchi, Atsuko

doi:10.1002/humu.24341

Cited by 4 publications

(3 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To perform better than phenotype-based gene prioritization methods, LLMs need to be able to replace these two main components. LLMs obtain their background knowledge from literature; the content of most genotypeto-phenotype databases will at least to large parts be reported in the literature (for example in the form of clinical case report [61]), and from our results, we can observe, similar to results in other clinical domains [24], that LLMs are as good or better in extracting the relevant information. LLMs also seem to be able to compute similarity between phenotypes as well or better than the custom-built similarity measures in Exomiser, demonstrated in particularly when using clinical phenotype descriptions as input to the ranking model (Table 5).…”

Section: Explanations Hallucinations and Reproducibilitysupporting

confidence: 60%

The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients

Kafkas,

Abdelhakim,

Althagafi

et al. 2023

Preprint

View full text Add to dashboard Cite

Computational methods for identifying gene–disease associations can use both genomic and phenotypic information to prioritize genes and variants that may be associated with genetic diseases. Phenotype-based methods commonly rely on comparing phenotypes observed in a patient with a database of genotype-to-phenotype associations using a measure of semantic similarity, and are primarily limited by the quality and completeness of this database as well as the quality of phenotypes assigned to a patient. Genotype-to-phenotype associations used by these methods are largely derived from literature and coded using phenotype ontologies. Large Language Models (LLMs) have been trained on large amounts of text and have shown their potential to answer complex questions across multiple domains. Here, we demonstrate that LLMs can prioritize disease-associated genes as well, or better than, dedicated bioinformatics methods relying on calculated phenotype similarity. The LLMs use only natural language information as background knowledge and do not require ontology-based phenotyping or structured genotype-to-phenotype knowledge. We use a cohort of undiagnosed patients with rare diseases and show that LLMs can be used to provide diagnostic support that helps in identifying plausible candidate genes.

show abstract

Section: Explanations Hallucinations and Reproducibilitysupporting

confidence: 60%

The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients

Kafkas,

Abdelhakim,

Althagafi

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The MME connects to three databases that provide additional utility to the gene discovery scientific community and these databases are considered "Connected Knowledge Sources," or "Functional Study Node," specialized MME endpoints that go beyond the initial MME design of two-sided genomic matchmaking (Figure 2). PubCaseFinder helps users identify any existing case reports for candidate genes by using phenotype-based comparisons (Fujiwara et al 2018(Fujiwara et al , 2022. To facilitate important downstream translational research using in vitro and in vivo models (Boycott et al, 2020;Wangler et al, 2017), MME connects to two additional databases; Monarch Initiative and ModelMatcher.…”

Section: Mme Supports Connections To Knowledge and Model Organism Res...mentioning

confidence: 99%

Seven years since the launch of the Matchmaker Exchange: The evolution of genomic matchmaking

et al. 2022

View full text Add to dashboard Cite

The Matchmaker Exchange (MME) was launched in 2015 to provide a robust mechanism to discover novel disease-gene relationships. It operates as a federated network connecting databases holding relevant data using a common application programming interface, where two or more users are looking for a match for the same gene (two-sided matchmaking). Seven years from its launch, it is clear that the MME is making outstanding contributions to understanding the morbid anatomy of the genome. The number of unique genes present across the MME has steadily increased over time; there are currently >13,520 unique genes (~68% of all proteincoding genes) connected across the MME's eight genomic matchmaking nodes, GeneMatcher, DECIPHER, PhenomeCentral, MyGene2, seqr, Initiative on Rare and Undiagnosed Disease, PatientMatcher, and the RD-Connect Genome-Phenome Analysis Platform. The collective data set accessible across the MME currently includes more than 120,000 cases from over 12,000 contributors in 98 countries. The discovery of potential new disease-gene relationships is happening daily and international collaborative teams are moving these advances forward to publication, now numbering well over 500. Expansion of data sharing into routine clinical practice by clinicians, genetic counselors, and clinical laboratories has ensured access to discovery for even more individuals with undiagnosed rare genetic diseases. Tens of thousands of patients and their family members have been directly or indirectly impacted by the discoveries facilitated by two-sided genomic matchmaking. MME supports further connections to the literature (PubCaseFinder) and to human and model organism resources (Monarch Initiative) and scientists (ModelMatcher).

show abstract

“…clinically-based interventions may start to offer support. 18 of the 51 papers which were included in our review aimed to support primary healthcare professionals with diagnosis decisions[47,60,63,66,[72][73][74][75]78,82,85,89,90,93,94,[103][104][105]. These works provided novel diagnostic solutions but did not provide support for referral decisions.…”

mentioning

confidence: 99%

Beyond Rare Disease Patients: Exploring Machine Learning Interventions To Support People Experiencing a Diagnostic Odyssey

Nielsen

View full text Add to dashboard Cite

People with a rare condition face several hurdles throughout their odyssey to obtain a diagnosis. This odyssey lasts several years and involves frequent referrals and misdiagnoses, often resulting in permanent and severe consequences on patients’ health. In addition, patients feel unheard by their healthcare providers and isolated from their peers who ‘just don’t understand’. The UK Strategy for Rare Diseases states that patients can play a significant role in their diagnosis if given suitable resources. However, patients with rare diseases feel that they lack the support they need. This thesis explores the role that technology can have in addressing this gap in support.Within this context, this thesis spans a range of topics, from human-centred design approaches to generating data and presenting a new methodological approach. Through a human-centred approach, we characterise the needs of rare disease patients, thus opening the research space to include previously unmet support needs. In addition, we identify limitations with existing measures of success and highlight the importance of a reduction in the time of diagnosis for rare disease pre-diagnostic technology. This provides the basis the simulation-based methodological approach that we develop. The simulation-task aimed to mirror the information seeking tasks that rare disease patients undertake. To do this, we curate data that is representative of a rare disease patient’s perspective, both in terms of the terminology used and the stage in which symptoms and clinical findings are discovered. In addition, we curate a pre-diagnostic patient matching prototype that is designed around rare disease patients’ needs and demonstrate that (in comparison to two search engines) our application shows greater potential to: aid clinical experiences; facilitate empathetic support networks; and provide better facilitation of information-seeking. All of these contributions stem from a critical examination of the experiences that rare disease patients go through on their journeys towards diagnosis and aim to pave the way for future research within this area.

show abstract

Advances in the development of PubCaseFinder, including the new application programming interface and matching algorithm

Cited by 4 publications

References 34 publications

The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients

The application of Large Language Models to the phenotype-based prioritization of causative genes in rare disease patients

Seven years since the launch of the Matchmaker Exchange: The evolution of genomic matchmaking

Beyond Rare Disease Patients: Exploring Machine Learning Interventions To Support People Experiencing a Diagnostic Odyssey

Contact Info

Product

Resources

About