Machine learning algorithm improved automated droplet classification of ddPCR for detection of BRAF V600E in paraffin-embedded samples

Abstract

Somatic mutations in cancer driver genes can help diagnosis, prognosis and treatment decisions. Formalin-fixed paraffin-embedded (FFPE) specimen is the main source of DNA for somatic mutation detection. To overcome constraints of DNA isolated from FFPE, we compared pyrosequencing and ddPCR analysis for absolute quantification of BRAF V600E mutation in the DNA extracted from FFPE specimens and compared the results to the qualitative detection information obtained by Sanger Sequencing. Sanger sequencing was able to detect BRAF V600E mutation only when it was present in more than 15% total alleles. Although the sensitivity of ddPCR is higher than that observed for Sanger, it was less consistent than pyrosequencing, likely due to droplet classification bias of FFPE-derived DNA. To address the droplet allocation bias in ddPCR analysis, we have compared different algorithms for automated droplet classification and next correlated these findings with those obtained from pyrosequencing. By examining the addition of non-classifiable droplets (rain) in ddPCR, it was possible to obtain better qualitative classification of droplets and better quantitative classification compared to no rain droplets, when considering pyrosequencing results. Notable, only the Machine learning k-NN algorithm was able to automatically classify the samples, surpassing manual classification based on no-template controls, which shows promise in clinical practice.

Diagnostic, Prognostic, and Therapeutic Value of Droplet Digital PCR (ddPCR) in COVID-19 Patients: A Systematic Review

“…Thus, optimization of the ddPCR assay and strict quality control measures are needed to achieve accurate and reproducible results [ 58 , 62 ]. Improved artificial intelligence algorithms could also help towards this direction [ 64 ].…”

Droplet Detection and Sorting System in Microfluidics: A Review

“…The absorbance of target-containing droplets has also been used as label-free detection signals which, therefore, can trigger the sorting of droplets ( Figure 3 c). Limited to optical path length, the sensitivity of absorbance measurements can be improved by the elongation of the channel, extension of residence time, or integration of other techniques such as fluorescence-based and UV–vis spectra-based techniques [ 122 , 123 ] Future development may extend to full absorbance spectra and integrate analysis by artificial intelligence [ 124 , 125 ]. Due to the advances in cameras, processing hardware, and machine learning software with image processing algorithms, bright field images of droplets have been able to be real-time captured and processed into signals for droplet sorting ( Figure 3 d) [ 126 ].…”

Comparative efficiency of differential diagnostic methods for the identification of BRAF V600E gene mutation in papillary thyroid cancer (Review)

“…Colozza-Gama et al ( 19 ) previously compared Sanger sequencing and pyrosequencing for detecting a somatic driver mutation, and observed that pyrosequencing was vastly superior for the detection of single nucleotide variants, particularly in highly degraded tumor samples derived from formalin-fixed paraffin-embedded (FFPE) specimens. Using DNA samples isolated from FFPE specimens, all papillary thyroid microcarcinoma and lymph node metastases samples were screened for BRAF V600E mutation by pyrosequencing.…”

“…In the era of next-generation sequencing (NGS) techniques, detection and analysis of the BRAF V600E mutation have been performed under clinical settings using a variety of different methods such as Sanger sequencing ( 18 ), pyrosequencing ( 19 ), reverse transcription-quantitative PCR (RT-qPCR) ( 20 ), amplification refractory mutation system (ARMS), NGS technology, high-resolution melting (HRM), droplet digital PCR (ddPCR) ( 21 ), MassArray ( 22 ) and immunohistochemistry (IHC)-based mutation detection ( 23 ). Among these methods, Sanger sequencing is considered to be the ‘gold standard’ in the majority of diagnostic studies.…”