SEQENS: An ensemble method for relevant gene identification in microarray data

Signol, François; Arnal, Laura; Navarro-Cerdan, J. Ramon; Llobet, Rafael; Arlandis, Joaquim; Pérez-Cortés, Juan-Carlos

doi:10.1016/j.compbiomed.2022.106413

Cited by 8 publications

(15 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cliff score is computed on such ranks. 1 These parameters were optimized empirically, as detailed in Signol et al (2023).…”

Section: Performance Comparison With Other Methodsmentioning

confidence: 99%

“…SEQENS (Signol et al, 2023) is an ensemble feature identification method whose kernel is a Sequential Feature Search (SFS) algorithm . As a wrapper method, SFS benefits SEQENS by identifying potential interactions among relevant features using inducers (or selectors), as well as, making no assumptions on variable distributions or their interaction types (independence, linearity, or any other kernels).…”

Section: Seqensmentioning

confidence: 99%

“…In Signol et al (2023), the multiple instances are obtained by means of splitting the whole input dataset into different training-test set partitions (stage b in Figure 3). This is one of the key advantages of the ensemble methods: multiple training data are taken to the selectors, which allows learning from different data distributions, which, in turn, leads to a bias reduction (Bolón-Canedo and Alonso-Betanzos, 2018) and stability improvement (Seijo-Pardo et al, 2017).…”

Section: Seqensmentioning

confidence: 99%

“…In this paper, we propose a framework based on SEQENS (Signol et al, 2023), a FID method that performs an ensemble of SFS (Sequential Feature Search), to improve SEQENS feature identification, which has demonstrated strong performance in identifying relevant features across different synthetic tasks. This framework spans the following contributions: (1) computing the hypergeometric p-value of the features of a SEQENS ensemble in order to threshold features into relevant and non-relevant;…”

mentioning

confidence: 99%

See 3 more Smart Citations

Feature identification using hypotheses of relevance and a 2D-cascade of SEQENS ensembles

Arlandis,

Llobet,

Navarro-Cerdán

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

SEQENS is an ensemble method aimed at feature identification that has demonstrated strong performance in identifying relevant features across different synthetic tasks (Signol et al, 2023). In this paper, we present a framework based on SEQENS spanning the following contributions: (1) computing the hypergeometric p-value of the features of a SEQENS output ranking in order to threshold into relevant and non-relevant features; (2) extending SEQENS by introducing the use of preselected features as hypotheses of relevance in the SFS, which may help to attract other features that might exhibit weak correlation with the target on their own, but gain relevance when combined with the preselected ones; (3) designing an automated process based on a 2D-cascade of SEQENS ensembles to obtain a \emph{purged feature set}, or PFS, i.e., having as many relevant features, and as few non-relevant, as possible; (4) integrating all the former techniques so that the PFS is used as in a SEQENS ensemble, which corresponds the complete framefork named pc-SEQENS. The performance of pc-SEQENS is measured on a task of gene expression identification using the E-MTAB-3732 public database and synthetic groundtruths. pc-SEQENS is compared to other feature identification state-of-the-art methods, including SEQENS. On average, the proposed framework identifies better the relevant genes, specially in the most unfavorable sample-to-dimension rates, and exhibits a stronger stability.

show abstract

“…Cliff score is computed on such ranks. 1 These parameters were optimized empirically, as detailed in Signol et al (2023).…”

Section: Performance Comparison With Other Methodsmentioning

confidence: 99%

Section: Seqensmentioning

confidence: 99%

Section: Seqensmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Feature identification using hypotheses of relevance and a 2D-cascade of SEQENS ensembles

Arlandis,

Llobet,

Navarro-Cerdán

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…A new line related to future T2D risk prediction is intended to be carried out by mixing environmental and lifestyle features with genomic data by selecting single-nucleotide polymorphism (SNPs) that appear to be related in any way to T2D disease. Further studies in this direction could take profit of genomic SNPs feature selection done in a previous work during the last years by applying innovative ensemble feature selection algorithm explained in [22]. The final result of the algorithm was an ordered list of the most voted features that are directed related to feature relevance concerning T2D, which allowed biologists to reduce cost during measuring part, focusing only on the most relevant features for new individuals study.…”

Section: Future Workmentioning

confidence: 99%

Machine-learning approach for Type 2 Diabetes diagnosis and prognosis models over heterogeneous feature spaces

Navarro-Cerdan

Pons-Suñer

Arnal

et al. 2023

Preprint

View full text Add to dashboard Cite

This research aims to evaluate the Type 2 Diabetes (T2D) diagnosis and prognosis power from heterogeneous environmental, lifestyle and biochemistry data. Model estimation has previously addressed three main actions as: 1) Missing-value imputation using specific univariant and multivariant imputers accommodated to each particular feature; 2) Quasi-constancy detection in variables; 3) Constructing geographical pollution and rent data from municipality information. Next, different T2D diagnosis and prognosis models are fitted and evaluated, showing increasing performance as more specific features become available while the prediction cost rises as a consequence of requiring more specific data. Finally, four models are obtained: two of them for T2D diagnosis and the other two for T2D prognosis respectively, with performances ranging from 73.3 to 95.41 AUC-ROC. One pair of diagnosis and prognosis models were thought for a global testing that can be done in general locations by only asking general lifestyle-related questions. On the other hand, the other pair, which achieves higher performances, is thought to be applied in a clinical environment where it is easy to obtain more specific biochemistry measures.

show abstract

An Ensemble Machine Learning Method Highlights Possible Parkinson’s Disease Genes and Accessing Performance of Re-sampling Techniques

Arora,

Mishra,

Malhi

2024

SN COMPUT. SCI.

View full text Add to dashboard Cite

SEQENS: An ensemble method for relevant gene identification in microarray data

Cited by 8 publications

References 51 publications

Feature identification using hypotheses of relevance and a 2D-cascade of SEQENS ensembles

Feature identification using hypotheses of relevance and a 2D-cascade of SEQENS ensembles

Machine-learning approach for Type 2 Diabetes diagnosis and prognosis models over heterogeneous feature spaces

An Ensemble Machine Learning Method Highlights Possible Parkinson’s Disease Genes and Accessing Performance of Re-sampling Techniques

Contact Info

Product

Resources

About