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.

The detection of ROS1 and ALK rearrangements is performed for advanced-stage non-small cell lung cancer. Several techniques can be used on cytological samples, such as immunocytochemistry (ICC), fluorescence in situ hybridization (FISH) and, more recently, next-generation sequencing (NGS), which is gradually becoming the gold standard. We performed a retrospective study to compare ALK and ROS1 rearrangement results from immunocytochemistry, FISH and NGS methods from 131 cytological samples. Compared to NGS, the sensitivity and specificity of ICC were 0.79 and 0.91, respectively, for ALK, and 1 and 0.87 for ROS1. Regarding FISH, the sensitivity and specificity were both at 1 for ALK and ROS1 probes. False-positive cases obtained by ICC were systematically corrected by FISH. When using ICC and FISH techniques, results are very close to NGS. The false-positive cases obtained by ICC are corrected by FISH, and the true-positive cases are confirmed. NGS has the potential to improve the detection of ALK and ROS1 rearrangements in cytological samples; however, the cost of this technique is still much higher than the sequential use of ICC and FISH.

The advent of next-generation sequencing (NGS) technologies has revolutionized the field of bioinformatics and genomics, particularly in the area of onco-somatic genetics. NGS has provided a wealth of information about the genetic changes that underlie cancer and has considerably improved our ability to diagnose and treat cancer. However, the large amount of data generated by NGS makes it difficult to interpret the variants. To address this, machine learning algorithms such as Extreme Gradient Boosting (XGBoost) have become increasingly important tools in the analysis of NGS data. In this paper, we present a machine learning tool that uses XGBoost to predict the pathogenicity of a mutation in the myeloid panel. We optimized the performance of XGBoost using metaheuristic algorithms and compared our predictions with the decisions of biologists and other prediction tools. The myeloid panel is a critical component in the diagnosis and treatment of myeloid neoplasms, and the sequencing of this panel allows for the identification of specific genetic mutations, enabling more accurate diagnoses and tailored treatment plans. We used datasets collected from our myeloid panel NGS analysis to train the XGBoost algorithm. It represents a data collection of 15,977 mutations variants composed of a collection of 13,221 Single Nucleotide Variants (SNVs), 73 Multiple Nucleoid Variants (MNVs), and 2683 insertion deletions (INDELs). The optimal XGBoost hyperparameters were found with Differential Evolution (DE), with an accuracy of 99.35%, precision of 98.70%, specificity of 98.71%, and sensitivity of 1.

Motivation, BRCA1 and BRCA2 are genes with tumor suppressor activity. They are involved in a considerable number of biological processes. To help the biologist in tumor classification, we developed a deep learning algorithm. The question when we want to construct a neural network is how many hidden layers and neurons should we use. If the number of inputs and outputs is defined by the problem, the number of hidden layers and neurons is difficult to define. Hidden layers and neurons that make up each layer of the neural network influence the performance of system predictions. There are different methods for finding the optimal architecture. In this paper, we present the two packages that we have developed, the genetic algorithm (GA) and the particle swarm optimization (PSO) to optimize the parameters of the neural network for predicting BRCA1 and BRCA2 pathogenicity; Results, we will compare the results obtained by the two algorithms. We used datasets collected from our NGS analysis of BRCA1 and BRCA2 genes to train deep learning models. It represents a data collection of 11,875 BRCA1 and BRCA2 variants. Our preliminary results show that the PSO provided the most significant architecture of hidden layers and the number of neurons compared to grid search and GA; Conclusions, the optimal architecture found by the PSO algorithm is composed of 6 hidden layers with 275 hidden nodes with an accuracy of 0.98, precision 0.99, recall 0.98, and a specificity of 0.99.

Motivation: Since 2017, we are using IonTorrent NGS platform in our hospital in order to diagnose cancer and treatment. Analysis variants at each run take us a longtime and we are still struggling with some variants which look correct on the first look at their metrics but found to be negative when we look further into them. Can any Machine Learning algorithm help us to classify NGS variant calling ? This has determined us to investigate which ML could fit to our NGS data and to develop a tool which can be implemented in Routine in order to help Biologists. Introduction: Nowadays, one of medicine challenges is processing a significant amount of data. It’s particularly true in molecular biology with the advantage of Next Generation Sequencing (NGS) for molecular tumor profile determination and treatment selection. In addition to bioinformatics pipelines, Artificial Intelligence (AI) can offer a very valuable help in analyzing. Generating sequencing data from patient DNA samples has become easy to perform in clinical trials. But analyzing the huge amount 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 skill and a panel of bioinformatics and biostatistics tools, in which artificial intelligence is now a success factor in developing future routine diagnostics. However, cancer genome complexity and technical artifacts make identification of real variants a challenge. We present a Machine Learning method to classify pathogenic Single Nucleotide Variants (SNVs), SNP (Single Nucleotide Polymorphism), MNVs (Multiple Nucleotide Variants), Insertion, Deletion detected by NGS from tumors specimens for Colorectal, Melanoma, Lung and Glioma cancer. Methods: We compared our NGS data to different machine learning algorithms using the 10-fold cross validation method and to neural networks (Deep Learning) in order to measure the performance of the different ML algorithms and determine which one is a valid model for confirming NGS variant calls in cancer diagnostic. 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 30 % remaining. The model offering the best accuracy was chosen and implemented in NGS analysis routine. The artificial intelligence was developed with R script language version 3.6.0. Results: We trained our model on 102011 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 Neural Network was able to achieve an accuracy of 98% and a 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 positive. After adding false positive in our training database and implementing our RF model in routine, our error rate was always < 0.5%. Conclusion: Our RF model shows excellent results for onco-somatic NGS interpretation and it could easily be implemented in other molecular biology laboratories. AI is taking an increasingly important place in molecular biomedical analysis and could be very helpful on processing of amount medical data. Neural Networks showed a good capacity in the classification of variants and in the future may be useful in the prediction of more complex variants.

BRCA1 and BRCA2 are genes with tumor suppressor activity, and they are involved ina considerable number of biological processes allowing the regulation of the cellreplication cycle. A mutation in one of these two genes has a significant probability ofcausing cancer. We have set up within the platform a machine learning algorithm basedon the random forest to predict pathogenicity in colorectal, melanoma, lung, and gliomacancer. but this algorithm has revealed its limits when we want to predict on morecomplex genes like BRCA1 and BRCA2. To help the biologist in the classification oftumors, we decided to develop a deep learning algorithm.The question we ask ourselves when we want to construct a neural network is howmany hidden layers and neurons should we use. If the number of inputs and outputs isdefined by the problem that we require to resolve, the number of hidden layers andneurons is difficult to define because there is no pre-established rule. The number ofhidden layers and neurons that make up each layer of the neural network has aninfluence on the performance of system predictions. There are different methods forfinding the optimal architecture like grid search or based on empirical equations. Allthese techniques can be very time-consuming. In this paper, we will present the twopackages that we have developed, the genetic algorithm (GA) and the particle swarmoptimization (PSO) to optimize the parameters of the neural network for the predictionof the pathogenicity of the BRCA1 and BRCA2 genes. We will compare the resultsobtained by the two algorithms. We used datasets collected from our NGS analysis ofBRCA1 and BRCA2 genes to train deep learning models. This represents a datacollection of 11,875 BRCA1 and BRCA2 variants (BRCA1 benign 2,632, BRCA1pathogenic 2,660, BRCA2 benign 3,446, BRCA2 pathogenic 3,137). Our preliminaryresults show that the PSO provided the most significant architecture in terms of hiddenlayers and the number of neurons compared to grid search and GA. The optimalarchitecture found by the PSO algorithm is composed of 6 hidden layers with 275 hiddennodes with an accuracy of 0.98, precision 0.99, recall 0.98, and a specificity of 0.99.