Semi-supervised learning for somatic variant calling and peptide identification in personalized cancer immunotherapy

Sherafat, Elham; Force, Jordan; Măndoiu, Ion

doi:10.1186/s12859-020-03813-x

Cited by 9 publications

(7 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A third option for timsTOF analysis is the de novo-based PEAKS software. In the current workflow, we used the robust search engine MSGF+ due to its advantages in peptidomics searches (Kim and Pevzner, 2014;Sherafat et al, 2020) and compatibility with our post-processing tools of choice (Silva et al, 2019). But the integration of more specialized search engines that omit data manipulation prior to searches like MaxQuant or MSFragger into the MS 2 ReScore pipeline in the future could further expand the information extracted from timsTOF data and advance neuropeptide and sORF-encoded peptide identification.…”

Section: Discussionmentioning

confidence: 99%

Ion Mobility Coupled to a Time-of-Flight Mass Analyzer Combined With Fragment Intensity Predictions Improves Identification of Classical Bioactive Peptides and Small Open Reading Frame-Encoded Peptides

Peeters

Baggerman

Gabriels

et al. 2021

Front. Cell Dev. Biol.

View full text Add to dashboard Cite

Bioactive peptides exhibit key roles in a wide variety of complex processes, such as regulation of body weight, learning, aging, and innate immune response. Next to the classical bioactive peptides, emerging from larger precursor proteins by specific proteolytic processing, a new class of peptides originating from small open reading frames (sORFs) have been recognized as important biological regulators. But their intrinsic properties, specific expression pattern and location on presumed non-coding regions have hindered the full characterization of the repertoire of bioactive peptides, despite their predominant role in various pathways. Although the development of peptidomics has offered the opportunity to study these peptides in vivo, it remains challenging to identify the full peptidome as the lack of cleavage enzyme specification and large search space complicates conventional database search approaches. In this study, we introduce a proteogenomics methodology using a new type of mass spectrometry instrument and the implementation of machine learning tools toward improved identification of potential bioactive peptides in the mouse brain. The application of trapped ion mobility spectrometry (tims) coupled to a time-of-flight mass analyzer (TOF) offers improved sensitivity, an enhanced peptide coverage, reduction in chemical noise and the reduced occurrence of chimeric spectra. Subsequent machine learning tools MS2PIP, predicting fragment ion intensities and DeepLC, predicting retention times, improve the database searching based on a large and comprehensive custom database containing both sORFs and alternative ORFs. Finally, the identification of peptides is further enhanced by applying the post-processing semi-supervised learning tool Percolator. Applying this workflow, the first peptidomics workflow combined with spectral intensity and retention time predictions, we identified a total of 167 predicted sORF-encoded peptides, of which 48 originating from presumed non-coding locations, next to 401 peptides from known neuropeptide precursors, linked to 66 annotated bioactive neuropeptides from within 22 different families. Additional PEAKS analysis expanded the pool of SEPs on presumed non-coding locations to 84, while an additional 204 peptides completed the list of peptides from neuropeptide precursors. Altogether, this study provides insights into a new robust pipeline that fuses technological advancements from different fields ensuring an improved coverage of the neuropeptidome in the mouse brain.

show abstract

Section: Discussionmentioning

confidence: 99%

Ion Mobility Coupled to a Time-of-Flight Mass Analyzer Combined With Fragment Intensity Predictions Improves Identification of Classical Bioactive Peptides and Small Open Reading Frame-Encoded Peptides

Peeters

Baggerman

Gabriels

et al. 2021

Front. Cell Dev. Biol.

View full text Add to dashboard Cite

show abstract

“…In recent years, computational modeling tools for peptides have advanced significantly [ 108 , 109 , 110 , 111 , 112 ]. Similar to mAbs, peptides exhibit large and chemically diverse binding interfaces, but they show better biodistribution and tissue penetration than mAbs.…”

Section: Binding Mode and Binding Affinity Prediction Of The Pd-1/pd-...mentioning

confidence: 99%

Computational Approaches Drive Developments in Immune-Oncology Therapies for PD-1/PD-L1 Immune Checkpoint Inhibitors

Sobral

Luz

Almeida

et al. 2023

IJMS

View full text Add to dashboard Cite

Computational approaches in immune-oncology therapies focus on using data-driven methods to identify potential immune targets and develop novel drug candidates. In particular, the search for PD-1/PD-L1 immune checkpoint inhibitors (ICIs) has enlivened the field, leveraging the use of cheminformatics and bioinformatics tools to analyze large datasets of molecules, gene expression and protein–protein interactions. Up to now, there is still an unmet clinical need for improved ICIs and reliable predictive biomarkers. In this review, we highlight the computational methodologies applied to discovering and developing PD-1/PD-L1 ICIs for improved cancer immunotherapies with a greater focus in the last five years. The use of computer-aided drug design structure- and ligand-based virtual screening processes, molecular docking, homology modeling and molecular dynamics simulations methodologies essential for successful drug discovery campaigns focusing on antibodies, peptides or small-molecule ICIs are addressed. A list of recent databases and web tools used in the context of cancer and immunotherapy has been compilated and made available, namely regarding a general scope, cancer and immunology. In summary, computational approaches have become valuable tools for discovering and developing ICIs. Despite significant progress, there is still a need for improved ICIs and biomarkers, and recent databases and web tools have been compiled to aid in this pursuit.

show abstract

“…Sherafat et al. ( 39 ) developed a positive-unlabeled learning model using auto machine learning to predict tumor-rejection mediation neoepitopes from exome sequencing data in ovarian cancer. The authors report improved performance over model-based classifiers for somatic variant calling and peptide identification.…”

Section: Studies On Semi-supervised Learning In Cancer Diagnosticsmentioning

confidence: 99%

Semi-supervised learning in cancer diagnostics

2022

View full text Add to dashboard Cite

In cancer diagnostics, a considerable amount of data is acquired during routine work-up. Recently, machine learning has been used to build classifiers that are tasked with cancer detection and aid in clinical decision-making. Most of these classifiers are based on supervised learning (SL) that needs time- and cost-intensive manual labeling of samples by medical experts for model training. Semi-supervised learning (SSL), however, works with only a fraction of labeled data by including unlabeled samples for information abstraction and thus can utilize the vast discrepancy between available labeled data and overall available data in cancer diagnostics. In this review, we provide a comprehensive overview of essential functionalities and assumptions of SSL and survey key studies with regard to cancer care differentiating between image-based and non-image-based applications. We highlight current state-of-the-art models in histopathology, radiology and radiotherapy, as well as genomics. Further, we discuss potential pitfalls in SSL study design such as discrepancies in data distributions and comparison to baseline SL models, and point out future directions for SSL in oncology. We believe well-designed SSL models to strongly contribute to computer-guided diagnostics in malignant disease by overcoming current hinderances in the form of sparse labeled and abundant unlabeled data.

show abstract

Semi-supervised learning for somatic variant calling and peptide identification in personalized cancer immunotherapy

Cited by 9 publications

References 32 publications

Ion Mobility Coupled to a Time-of-Flight Mass Analyzer Combined With Fragment Intensity Predictions Improves Identification of Classical Bioactive Peptides and Small Open Reading Frame-Encoded Peptides

Ion Mobility Coupled to a Time-of-Flight Mass Analyzer Combined With Fragment Intensity Predictions Improves Identification of Classical Bioactive Peptides and Small Open Reading Frame-Encoded Peptides

Computational Approaches Drive Developments in Immune-Oncology Therapies for PD-1/PD-L1 Immune Checkpoint Inhibitors

Semi-supervised learning in cancer diagnostics

Contact Info

Product

Resources

About