Systems and methods for detecting CRISPR-mediated residues within methylated patterns of genome using a convolutional neural network
A system, method, and computer-readable medium for detecting a CRISPR-edited genome are disclosed. Certain embodiments of the system may include one or more processors configured to receive sequence data of a genome; generate an image representation of the sequenced data, the image being a plot of methylation variations as a function of methylation locations in the genome; apply the generated image representation to a trained convoluted neural network (CNN); generate, using the CNN, a score indicative of a probability that the genome was CRISPR-edited; and determine, based on the score, whether the genome contains a CRISPR-edited methylation region. A corresponding method and computer-readable medium are also provided.
1 . A system for detecting a CRISPR-edited genome, comprising:
One or more processors configured to:
receive sequence data of a genome;
generate an image representation of the sequenced data, the image being a plot of methylation variations as a function of methylation locations in the genome;
provide the generated image representation to a convoluted neural network (CNN), wherein:
the CNN was trained using: (1) a training image representing sequenced data of a CRISPR-edited genome, and (2) an additional training image generated based on the training image, and
the additional training image is generated by (1) duplicating the training image, and (2) replicating a part of the training image corresponding to a CRISPR-edited methylation region of the CRISPR-edited genome to a part of the duplicated training image corresponding to a different methylation region;
generate, using the CNN, a score indicative of a probability that the genome was CRISPR-edited; and
determine, based on the score, whether the genome contains a CRISPR-edited methylation region.
2 . The system of claim 1 , wherein the sequenced data is whole-genome bisulfite sequencing (WGBS) data.
3 . The system of claim 1 , wherein the image is a Manhattan plot.
4 . The system of claim 1 , wherein the CNN includes convolution layers and a pooling layer.
5 . The system of claim 1 , wherein, after determining that the genome contains a CRISPR-edited methylation site, uses the image representation of the sequenced data to further train the CNN.
6 . The system of claim 1 , wherein the determining includes determining a methylation location of the CRISPR edit.
7 . The system of claim 1 , wherein one or more processors are further configured to determine, using an LSTM, whether the genome likely contains a CRISPR-edited methylation site, wherein the determining includes weighing the generated score and results from the LSTM to determine whether the genome contains a CRISPR edit.
8 . A method for detecting a CRISPR-edited genome, comprising:
receiving sequence data of a genome;
generating an image representation of the sequenced data, the image being a plot of methylation variations as a function of methylation locations in the genome;
providing the generated image representation to a convoluted neural network (CNN), wherein:
the CNN was trained using: (1) a training image representing sequenced data of a CRISPR-edited genome, and (2) an additional training image generated based on the training image, and
the additional training image is generated by (1) duplicating the training image, and (2) replicating a part of the training image corresponding to a CRISPR-edited methylation region of the CRISPR-edited genome to a part of the duplicated training image corresponding to a different methylation region;
generating, using the CNN, a score indicative of a probability that the genome was CRISPR-edited; and
determining, based on the score, whether the genome contains a CRISPR-edited methylation region.
9 . The method of claim 8 , wherein the sequenced data is whole-genome bisulfite sequencing (WGBS) data.
10 . The method of claim 8 , wherein the image is a Manhattan plot.
11 . The method of claim 8 , wherein the CNN includes convolution layers and a pooling layer.
12 . The method of claim 8 , wherein, after determining that the genome contains a CRISPR-edited methylation site, uses the image representation of the sequenced data to further train the CNN.
13 . The method of claim 8 , wherein the determining includes determining a methylation location of the CRISPR edit.
14 . The method of claim 8 , further comprising determining, using an LSTM, whether the genome likely contains a CRISPR-edited methylation site, wherein the determining includes weighing the generated score and results from the LSTM to determine whether the genome contains a CRISPR edit.
15 . A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for detecting a CRISPR-edited genome, the method comprising:
receiving sequence data of a genome;
generating an image representation of the sequenced data, the image being a plot of methylation variations as a function of methylation locations in the genome;
providing the generated image representation to a convoluted neural network (CNN), wherein:
the CNN was trained using: (1) a training image representing sequenced data of a CRISPR-edited genome, and (2) an additional training image generated based on the training image, and
the additional training image is generated by (1) duplicating the training image, and (2) replicating a part of the training image corresponding to a CRISPR-edited methylation region of the CRISPR-edited genome to a part of the duplicated training image corresponding to a different methylation region;
generating, using the CNN, a score indicative of a probability that the genome was CRISPR-edited; and
determining, based on the score, whether the genome contains a CRISPR-edited methylation region.
16 . The non-transitory computer-readable storage medium according to claim 15 , wherein the sequenced data is whole-genome bisulfite sequencing (WGBS) data.
17 . The non-transitory computer-readable storage medium according to claim 15 , wherein the image is a Manhattan plot.
18 . The non-transitory computer-readable storage medium according to claim 15 , wherein the CNN includes convolution layers and a pooling layer.
19 . The non-transitory computer-readable storage medium according to claim 15 , wherein, after determining that the genome contains a CRISPR-edited methylation site, uses the image representation of the sequenced data to further train the CNN.
20 . The non-transitory computer-readable storage medium according to claim 15 , wherein the determining includes determining a methylation location of the CRISPR edit.