Machine learning random forest for predicting oncosomatic variant NGS analysis.

Since 2017, we have used IonTorrent NGS platform in our hospital to diagnose and treat cancer. Analyzing variants at each run requires considerable time, and we are still struggling with some variants that appear correct on the metrics at first, but are found to be negative upon further investigation. Can any machine learning algorithm (ML) help us classify NGS variants? This has led us to investigate which ML can fit our NGS data and to develop a tool that can be routinely implemented to help biologists. Currently, one of the greatest challenges in medicine is processing a significant quantity of data. This is particularly true in molecular biology with the advantage of next-generation sequencing (NGS) for profiling and identifying molecular tumors and their treatment. In addition to bioinformatics pipelines, artificial intelligence (AI) can be valuable in helping to analyze mutation variants. Generating sequencing data from patient DNA samples has become easy to perform in clinical trials. However, analyzing the massive quantities of genomic or transcriptomic data and extracting the key biomarkers associated with a clinical response to a specific therapy requires a formidable combination of scientific expertise, biomolecular skills and a panel of bioinformatic and biostatistic tools, in which artificial intelligence is now successful in developing future routine diagnostics. However, cancer genome complexity and technical artifacts make identifying real variants challenging. We present a machine learning method for classifying pathogenic single nucleotide variants (SNVs), single nucleotide polymorphisms (SNPs), multiple nucleotide variants (MNVs), insertions, and deletions detected by NGS from different types of tumor specimens, such as: colorectal, melanoma, lung and glioma cancer. We compared our NGS data to different machine learning algorithms using the k-fold cross-validation method and to neural networks (deep learning) to measure the performance of the different ML algorithms and determine which one is a valid model for confirming NGS variant calls in cancer diagnosis. We trained our machine learning with 70% of our data samples, extracted from our local database (our data structure had 7 parameters: chromosome, position, exon, variant allele frequency, minor allele frequency, coverage and protein description) and validated it with the 30% remaining data. The model offering the best accuracy was chosen and implemented in the NGS analysis routine. Artificial intelligence was developed with the R script language version 3.6.0. We trained our model on 70% of 102,011 variants. Our best error rate (0.22%) was found with random forest machine learning (ntree = 500 and mtry = 4), with an AUC of 0.99. Neural networks achieved some good scores. The final trained model with the neural network achieved an accuracy of 98% and an ROC-AUC of 0.99 with validation data. We tested our RF model to interpret more than 2000 variants from our NGS database: 20 variants were misclassified (error rate < 1%). The errors were nomenclature problems and false positives. After adding false positives to our training database and implementing our RF model routinely, our error rate was always < 0.5%. The RF model shows excellent results for oncosomatic NGS interpretation and can easily be implemented in other molecular biology laboratories. AI is becoming increasingly important in molecular biomedical analysis and can be very helpful in processing medical data. Neural networks show a good capacity in variant classification, and in the future, they may be useful in predicting more complex variants.

Identification of potential prognostic biomarkers for node-negative breast tumours by proteomic analysis: a multicentric 2004 national PHRC study.

We used a 2D-electrophoresis (2-DE) proteomic approach to identify novel biomarkers in node-negative breast cancers. This retrospective study focused on a population of patients with ductal pN0M0 tumours. A subset of patients who developed metastases and in whose tumours were found high levels of uPA and PAI-1 (metastatic relapse, MR: n=20) were compared to another subset in whom no metastatic relapse occurred and whose tumours were found to have low levels of uPA and PAI-1 (no relapse, NR: n=21). We used a 2-DE coupled with MS approach to screen cytosol fractions using two pH-gradient scales, a broad scale (3.0-11.0) and a narrower scale focussing in on a protein rich region (5.0-8.0). This study was conducted on 41 cytosol specimens analyzed in duplicate on two platforms. The differential analysis of more than 2,000 spots in 2-DE gels, obtained on the two platforms, allowed the identification of 13 proteins which were confirmed by western blotting. Two proteins, GPDA and FABP4 were down-regulated in the MR subset whereas all the others were up-regulated. An in silico analysis revealed that GMPS (GUAA), GAPDH (G3P), CFL1 (COF1) and FTL (FRIL), the most informative genes, displayed a proliferation profile (high expression in basal-like, HER2+ and luminal B molecular subtypes). Inversely, similar to FABP4, GPD1 [GPDA] displayed a high expression in luminal A subtype, a profile characteristic of tumour suppressor genes. Despite the small size of our cohort, the 2-DE analysis gave interesting results which were confirmed by the in silico analysis showing that some of the corresponding genes had a strong prognostic impact in breast cancer, mostly because of their link with proliferation: GMPS, GAPDH, FTL and GPD1. A validation phase on a larger cohort is now needed before these biomarkers could be considered for use in clinical practice.

Validation of tumor-associated macrophage ferritin light chain as a prognostic biomarker in node-negative breast cancer tumors: A multicentric 2004 national PHRC study.

Novel prognostic biomarkers are imperatively needed to help direct treatment decisions by typing subgroups of node-negative breast cancer patients. Large screening of different biological compartments, such as the proteome, by means of high throughput techniques may greatly help scientists to find such markers. The present retrospective multicentric study included 268 node-negative breast cancer patients. We used a proteomic approach of SELDI-TOF-MS screening to identify differentially expressed cytosolic proteins with prognostic impact. The screening cohort was composed of 198 patients. Seventy supplementary patients were included for validation. Immunohistochemistry (IHC) and immunoassay (IA) were run to confirm the prognostic role of the marker identified by SELDI-TOF-MS screening. IHC was also used to explore links between selected marker and epithelial-mesenchymal transition (EMT)-like, proliferation and macrophage markers. Ferritin light chain (FTL) was identified as an independent prognostic marker (HR = 1.30-95% CI: 1.10-1.50, p = 0.001). Validation step by means of IHC and IA confirmed the prognostic value of FTL level. CD68 IHC showed that FTL was stored in tumor-associated macrophages (TAM), which exhibit an M2-like phenotype. We report here, first, the validation of FTL as a breast tumor prognostic biomarker in node-negative patients, and second, the fact that FTL is stored in TAM.

Is there such a thing as Latin bioethics?

This paper reflects on the presumption that there are distinct ethical differences between the supposedly 'Anglo-Saxon liberal' and 'Latin (Southern European) paternalist' ethical traditions. The predominance of the bioethical paradigm (principalism) is measured by a comparative analysis of regional moral opinion reflected in nation-state health laws. By looking at the way the ethico-legal concept figures into various national ordinances, we attempt to ascertain the extent and nature of variation (if any) between localities by exploring the understanding and application of principalism's keystone: patient autonomy.