Isoelectric point barcode and similarity analysis with the earth mover’s distance for identification of species origin of raw meat

Liu, Tian; Tian, Youli; Cao, Yiren; Wang, Zihao; Zha, Genhan; Liu, Weiwen; Li, Wei; Xiao, Hua; Zhang, Qiang; Cao, Cheng-Xi

doi:10.1016/j.foodres.2023.112600

Cited by 4 publications

(11 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The mIEF method is advantageous over the slab gel IEF for meat sample identification due to its higher throughput and shorter assay time. 8,13,18 For a 10 min mIEF run, 24 microstrips were placed on the mIEF array for parallel testing (Figures 1A and S1). The focusing bands of two pI markers, Hb, Mb, and oxygen-binding variants were imaged to produce an IEF pattern specific to a meat species.…”

Section: ■ Results and Discussionmentioning

confidence: 99%

“…The focusing bands of two pI markers, Hb, Mb, and oxygen-binding variants were imaged to produce an IEF pattern specific to a meat species. We previously utilized a similarity analysis algorithm, EMD, to scan the IEF patterns . However, we must conduct pre-experiments for optimizing the selection of pI markers, so as to cover the pI values of meat proteins.…”

Section: Resultsmentioning

confidence: 99%

“…The reason is that the EMD algorithm performed the identification according to the binarized pI barcode (Figure S2A), which did not provide any information about the grayscale value (i.e., peak area). The EMD method in a previous work could have achieved a high accuracy of 100% due to the use of the same cut of meat (only leg). However, the EMD method could only achieve a low accuracy of 76% due to the use of three/four cuts of meat with different grayscale values (Figures E,F).…”

Section: Resultsmentioning

confidence: 99%

“…Notably, the efficiency of the high-throughput and rapid mIEF platform is compromised because of the trivial procedure of locating pI markers and labeling the pI value of each protein band. 18 Herein, we proposed a marker-free IEF method to obviate the need for selecting suitable pI markers and labeling pI values, thus achieving high efficiency and large-scale analyses. To fulfill these improvements, we for the first time introduced convolutional neural network (CNN) to the IEF technique for the classification of IEF patterns.…”

Section: ■ Introductionmentioning

confidence: 99%

“…Thanks to the high throughput and short assay time of mIEF, 8,13,18 we easily collected 1564 images that recorded the patterns of meat Hb, myoglobin (Mb), and their oxygenbinding variants. 13,18 We aimed to accurately and rapidly differentiate cuts of meat (e.g., breast and leg meat) and meat species (e.g., cattle and sheep) at the same time. To the best of our knowledge, this study is the first example of using the IEF method without pI information for analyzing complicated protein samples.…”

Section: ■ Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Marker-Free Isoelectric Focusing Patterns for Identification of Meat Samples via Deep Learning

Tian,

Cao,

Zha

et al. 2023

Anal. Chem.

Self Cite

View full text Add to dashboard Cite

Isoelectric focusing (IEF) is a powerful tool for resolving complex protein samples, which generates IEF patterns consisting of multiplex analyte bands. However, the interpretation of IEF patterns requires the careful selection of isoelectric point (pI) markers for profiling the pH gradient and a trivial process of pI labeling, resulting in low IEF efficiency. Here, we for the first time proposed a marker-free IEF method for the efficient and accurate classification of IEF patterns by using a convolutional neural network (CNN) model. To verify our method, we identified 21 meat samples whose IEF patterns comprised different bands of meat hemoglobin, myoglobin, and their oxygen-binding variants but no pI marker. Thanks to the high throughput and short assay time of the microstrip IEF, we efficiently collected 1449 IEF patterns to construct the data set for model training. Despite the absence of pI markers, we experimentally introduced the severe pH gradient drift into 189 IEF patterns in the data set, thereby omitting the need for profiling the pH gradient. To enhance the model robustness, we further employed data augmentation during the model training to mimic pH gradient drift. With the advantages of simple preprocessing, a rapid inference of 50 ms, and a high accuracy of 97.1%, the CNN model outperformed the traditional algorithm for simultaneously identifying meat species and cuts of meat of 105 IEF patterns, suggesting its great potential of being combined with microstrip IEF for large-scale IEF analyses of complicated protein samples.

show abstract

Section: ■ Results and Discussionmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: ■ Introductionmentioning

confidence: 99%