2021
DOI: 10.1016/j.isci.2021.102484
|View full text |Cite
|
Sign up to set email alerts
|

Integrating protein networks and machine learning for disease stratification in the Hereditary Spastic Paraplegias

Abstract: A bioinformatic study of the Hereditary Spastic Paraplegias using protein networks.Human and manually curated protein-protein interaction data acquired using PINOT.Intracellular transport and vesicle trafficking are suggested as disease mechanisms.Machine learning techniques propose a patient clustering.

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
6
0

Year Published

2021
2021
2023
2023

Publication Types

Select...
5
1
1

Relationship

1
6

Authors

Journals

citations
Cited by 8 publications
(6 citation statements)
references
References 70 publications
0
6
0
Order By: Relevance
“…Recently, there has been an increasing tendency to involve the underlying molecular pathogenetic mechanisms in the categorization of HSP with multiple trials to establish classifications and groupings of identified HSP forms/genes based on the affected function (Lo Giudice et al, 2014;Noreau et al, 2014;Vallat et al, 2016;de Souza et al, 2017;Blackstone, 2018;Boutry et al, 2019a;. More recently, machine-learning approaches were also used in their clinico-genetic stratification (Vavouraki et al, 2021). This group of diseases suffers from the fact that it gathers diseases in which spasticity/pyramidal signs are sometimes only part of the clinical phenotype such as some neurodevelopmental disorders due to AP4n genes mutations, and, at the other end of the spectrum, disorders in which the spasticity/pyramidal signs are the core and sometimes the only symptoms in patients, such as most SPG4 and SPG8 patients.…”
Section: Classification Of Hspmentioning
confidence: 99%
“…Recently, there has been an increasing tendency to involve the underlying molecular pathogenetic mechanisms in the categorization of HSP with multiple trials to establish classifications and groupings of identified HSP forms/genes based on the affected function (Lo Giudice et al, 2014;Noreau et al, 2014;Vallat et al, 2016;de Souza et al, 2017;Blackstone, 2018;Boutry et al, 2019a;. More recently, machine-learning approaches were also used in their clinico-genetic stratification (Vavouraki et al, 2021). This group of diseases suffers from the fact that it gathers diseases in which spasticity/pyramidal signs are sometimes only part of the clinical phenotype such as some neurodevelopmental disorders due to AP4n genes mutations, and, at the other end of the spectrum, disorders in which the spasticity/pyramidal signs are the core and sometimes the only symptoms in patients, such as most SPG4 and SPG8 patients.…”
Section: Classification Of Hspmentioning
confidence: 99%
“…The semantic classes were grouped further in functional groups. The functional groups named General and Metabolism were not included in the analysis, as done previously (33). Neuronal-related terms were observed in multiple functional groups, so text mining using the keys “nerv”, “neur”, “brai”, and “synap”, was used to identify these terms in our results.…”
Section: Methodsmentioning
confidence: 99%
“…Neuronal-related terms were observed in multiple functional groups, so text mining using the keys “nerv”, “neur”, “brai”, and “synap”, was used to identify these terms in our results. The calculation of the enrichment ratios of the neuron-related words and the statistical analysis were calculated, as previously described (33). The p value was calculated using the Excel function: =2*(1-NORM.DIST(x, mean,sd,TRUE)).…”
Section: Methodsmentioning
confidence: 99%
“…Recent advancements in artificial intelligence and machine learning have provided the means to utilise network analysis tools for the stratifying disease features in HSP, offering novel approaches to enhance our understanding of HSP subtypes. 7 By employing bioinformatics tools alongside molecular studies, this approach has revealed shared biological processes among clusters of HSP patients. Although further research is required to replicate and validate these findings, they suggest that patients can be grouped into distinct aetiological categories, based on molecular features that characterising HSP in each case.…”
Section: Introductionmentioning
confidence: 99%