Predictive data analysis using image representations of genomic data
There is a need for more effective and efficient predictive data analysis solutions and/or more effective and efficient solutions for generating image representations of genetic variant data. In one example, embodiments comprise receiving an input feature, generating one or more image representations of the input feature, generating a tensor representation of the one or more image representations, generating a plurality of positional encoding maps, generating an image-based prediction based at least in part on the image representation, and performing one or more prediction-based actions based at least in part on the image-based prediction.
1. A computer-implemented method for dynamically generating an image-based prediction, the computer-implemented method comprising:
receiving, by one or more processors, an input feature, wherein the input feature comprises one or more feature values, wherein each feature value of the one or more feature values corresponds to a genetic variant identifier of a plurality of genetic variant identifiers, and wherein each feature value of the one or more feature values is associated with an input feature type designation of a plurality of input feature type designations;
generating, by the one or more processors, one or more image representations of the input feature, wherein: (i) an image representation count of the one or more image representations of the input feature is based at least in part on the plurality of input feature type designations (ii) each image representation of the one or more image representations of the input feature comprises a plurality of image regions, (iii) each image region of the plurality of image regions for an image representation of the one or more image representations of the input feature corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, and (iv) generating each of the one or more image representations of the input feature associated with a character category is performed based at least in part on the one or more feature values of the input feature having the input feature type designation;
generating, by the one or more processors, a tensor representation of the one or more image representations of the input feature;
generating, by the one or more processors, one or more positional encoding maps, wherein: (i) each positional encoding map of the one or more positional encoding maps comprises a plurality of positional encoding map regions, (ii) each positional encoding map region of the plurality of positional encoding map regions for a positional encoding map corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, (iii) each specific genetic variant identifier of the plurality of genetic variant identifiers is associated with a positional encoding map region set, of one or more positional encoding map region sets, comprising each positional encoding map region of the plurality of positional encoding map regions associated with the specific genetic variant identifier across the one or more positional encoding maps, and (iv) each positional encoding map region set of the one or more positional encoding map region sets for a particular genetic variant identifier of the plurality of genetic variant identifiers represents the particular genetic variant identifier;
generating, by the one or more processors, the image-based prediction based at least in part on the tensor representation of the one or more image representations of the input feature and the one or more positional encoding maps; and
performing, by the one or more processors, one or more prediction-based actions based at least in part on the image-based prediction.
2. The computer-implemented method of claim 1 , wherein generating the one or more image representations of the input feature further comprises:
generating, by the one or more processors, a first image representation generated based at least in part on a first subset of input features;
generating, by the one or more processors, a second image representation generated based at least in part on a second subset of input features; and
generating, by the one or more processors, a differential image representation of the one or more image representations of the input feature based at least in part on performing an image difference operation across the first image representation and the second image representation.
3. The computer-implemented method of claim 1 , wherein generating the one or more image representations of the input feature further comprises:
generating, by the one or more processors, a first allele image representation generated based at least in part on a first subset of input features corresponding to a first allele;
generating, by the one or more processors, a second allele image representation generated based at least in part on a second subset of input features corresponding to a second allele;
generating, by the one or more processors, a dominant allele image representation generated based at least in part on a dominant subset of input features corresponding to a dominant allele;
generating, by the one or more processors, a minor allele image representation generated based at least in part on a minor subset of input features corresponding to a minor allele; and
generating, by the one or more processors, a zygosity image representation of the one or more image representations of the input feature based at least in part on performing one or more operations across the first allele image representation, the second allele image representation, the dominant allele image representation, and the minor allele image representation.
4. The computer-implemented method of claim 1 , wherein generating the one or more image representations of the input feature further comprises:
identifying one or more initial image representations of the input feature;
assigning, by the one or more processors, one or more intensity values to each input feature type designation of the plurality of input feature type designations;
generating, by the one or more processors, one or more intensity image representations of the one or more initial image representations, wherein (i) each image representation of the one or more intensity image representations comprises a plurality of intensity image regions, (ii) each image region of the plurality of intensity image regions for an intensity image representation corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, and (iii) generating the one or more intensity image representations is determined based at least in part on the one or more feature values and the one or more assigned intensity value for each input feature type designation of the plurality of input feature type designations.
5. The computer-implemented method of claim 1 , wherein the image-based prediction comprises generating, by the one or more processors, a polygenic risk score for one or more diseases for one or more individuals associated with the input feature.
6. The computer-implemented method of claim 1 , wherein each feature value of the one or more feature values corresponds to a categorical feature type or numerical feature type.
7. The computer-implemented method of claim 1 , wherein each feature value of the one or more feature values further corresponds to a chromosome number and locus.
8. A system for dynamically generating an image-based prediction, the system comprising one or more processors and memory including program code, the memory and the program code configured to, with the one or more processors, cause the system to at least:
receive an input feature, wherein the input feature comprises one or more feature values, wherein each feature value of the one or more feature values corresponds to a genetic variant identifier of a plurality of genetic variant identifiers, and wherein each feature value of the one or more feature values is associated with an input feature type designation of a plurality of input feature type designations;
generate one or more image representations of the input feature, wherein: (i) an image representation count of the one or more image representations of the input feature is based at least in part on the plurality of input feature type designations (ii) each image representation of the one or more image representations of the input feature comprises a plurality of image regions, (iii) each image region of the plurality of image regions for an image representation of the one or more image representations of the input feature corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, and (iv) generating each of the one or more image representations of the input feature associated with a character category is performed based at least in part on the one or more feature values of the input feature having the input feature type designation;
generate a tensor representation of the one or more image representations of the input feature;
generate one or more positional encoding maps, wherein: (i) each positional encoding map of the one or more positional encoding maps comprises a plurality of positional encoding map regions, (ii) each positional encoding map region of the plurality of positional encoding map regions for a positional encoding map corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, (iii) each specific genetic variant identifier of the plurality of genetic variant identifiers is associated with a positional encoding map region set, of one or more positional encoding map region sets, comprising each positional encoding map region of the plurality of positional encoding map regions associated with the specific genetic variant identifier across the one or more positional encoding maps, and (iv) each positional encoding map region set of the one or more positional encoding map region sets for a particular genetic variant identifier of the plurality of genetic variant identifiers represents the particular genetic variant identifier;
generate the image-based prediction based at least in part on the tensor representation of the one or more image representations of the input feature and the one or more positional encoding maps; and
perform one or more prediction-based actions based at least in part on the image-based prediction.
9. The system of claim 8 , the memory and the program code are further configured to, with the one or more processors, cause the system to:
generate a first image representation generated based at least in part on a first subset of input features;
generate a second image representation generated based at least in part on a second subset of input features; and
generate a differential image representation of the one or more image representations of the input feature based at least in part on performing an image difference operation across the first image representation and the second image representation.
10. The system of claim 8 , the memory and the program code are further configured to, with the one or more processors, cause the system to:
generate a first allele image representation generated based at least in part on a first subset of input features corresponding to a first allele;
generate a second allele image representation generated based at least in part on a second subset of input features corresponding to a second allele;
generate a dominant allele image representation generated based at least in part on a dominant subset of input features corresponding to a dominant allele;
generate a minor allele image representation generated based at least in part on a minor subset of input features corresponding to a minor allele;
generate a zygosity image representation of the one or more image representations of the input feature based at least in part on performing one or more operations across the first allele image representation, the second allele image representation, the dominant allele image representation, and the minor allele image representation.
11. The system of claim 8 , the memory and the program code are further configured to, with the one or more processors, cause the system to:
identify one or more initial image representations of the input feature;
assign one or more intensity values to each input feature type designation of the plurality of input feature type designations;
generate one or more intensity image representations of the one or more initial image representations, wherein (i) each image representation of the one or more intensity image representations comprises a plurality of intensity image regions, (ii) each image region of the plurality of intensity image regions for an intensity image representation corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, and (iii) generating the one or more intensity image representations is determined based at least in part on the one or more feature values and the one or more assigned intensity value for each input feature type designation of the plurality of input feature type designations.
12. The system of claim 8 , wherein to generate the image-based prediction, the memory and the program code are further configured to, with the one or more processors, cause the system to generate a polygenic risk score for one or more diseases for one or more individuals associated with the input feature.
13. The system of claim 8 , wherein each feature value of the one or more feature values corresponds to a categorical feature type or numerical feature type.
14. The system of claim 8 , wherein each feature value of the one or more feature values further corresponds to a chromosome number and locus.
15. A computer program product for dynamically generating an image-based prediction, the computer program product comprising at least one non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions configured to:
receive an input feature, wherein the input feature comprises one or more feature values, wherein each feature value of the one or more feature values corresponds to a genetic variant identifier of a plurality of genetic variant identifiers, and wherein each feature value of the one or more feature values is associated with an input feature type designation of a plurality of input feature type designations;
generate one or more image representations of the input feature, wherein: (i) an image representation count of the one or more image representations of the input feature is based at least in part on the plurality of input feature type designations (ii) each image representation of the one or more image representations of the input feature comprises a plurality of image regions, (iii) each image region of the plurality of image regions for an image representation of the one or more image representations of the input feature corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, and (iv) generating each of the one or more image representations of the input feature associated with a character category is performed based at least in part on the one or more feature values of the input feature having the input feature type designation;
generate a tensor representation of the one or more image representations of the input feature;
generate one or more positional encoding maps, wherein: (i) each positional encoding map of the one or more positional encoding maps comprises a plurality of positional encoding map regions, (ii) each positional encoding map region of the plurality of positional encoding map regions for a positional encoding map corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, (iii) each specific genetic variant identifier of the plurality of genetic variant identifiers is associated with a positional encoding map region set, of one or more positional encoding map region sets, comprising each positional encoding map region of the plurality of positional encoding map regions associated with the specific genetic variant identifier across the one or more positional encoding maps, and (iv) each positional encoding map region set of the one or more positional encoding map region sets for a particular genetic variant identifier of the plurality of genetic variant identifiers represents the particular genetic variant identifier;
generate the image-based prediction based at least in part on the tensor representation of the one or more image representations of the input feature and the one or more positional encoding maps; and
perform one or more prediction-based actions based at least in part on the image-based prediction.
16. The computer program product of claim 15 , wherein the computer-readable program code portions is further configured to:
generate a first image representation generated based at least in part on a first subset of input features;
generate a second image representation generated based at least in part on a second subset of input features; and
generate a differential image representation of the one or more image representations of the input feature based at least in part on performing an image difference operation across the first image representation and the second image representation.
17. The computer program product of claim 15 , wherein the computer-readable program code portions is further configured to:
generate a first allele image representation generated based at least in part on a first subset of input features corresponding to a first allele;
generate a second allele image representation generated based at least in part on a second subset of input features corresponding to a second allele;
generate a dominant allele image representation generated based at least in part on a dominant subset of input features corresponding to a dominant allele;
generate a minor allele image representation generated based at least in part on a minor subset of input features corresponding to a minor allele;
generate a zygosity image representation of the one or more image representations of the input feature based at least in part on performing one or more operations across the first allele image representation, the second allele image representation, the dominant allele image representation, and the minor allele image representation.
18. The computer program product of claim 15 , wherein the computer-readable program code portions is further configured to:
identify one or more initial image representations of the input feature;
assign one or more intensity values to each input feature type designation of the plurality of input feature type designations;
generate one or more intensity image representations of the one or more initial image representations, wherein (i) each image representation of the one or more intensity image representations comprises a plurality of intensity image regions, (ii) each image region of the plurality of intensity image regions for an intensity image representation corresponds to a genetic variant identifier of the plurality of genetic variant identifiers, and (iii) generating the one or more intensity image representations is determined based at least in part on the one or more feature values and the one or more assigned intensity value for each input feature type designation of the plurality of input feature type designations.
19. The computer program product of claim 15 , wherein the image-based prediction comprises generating a polygenic risk score for one or more diseases for one or more individuals associated with the input feature.
20. The computer program product of claim 15 , wherein each feature value of the one or more feature values corresponds to a categorical feature type or numerical feature type.