WHITAKER JOHN W (US)
WSGR Docket No.44503-747.601 CLAIMS WHAT IS CLAIMED IS: 1. A method for determining if a subject has atopic dermatitis or psoriasis, comprising: receiving a plurality of expression levels, wherein the expression levels are derived from a skin sample obtained using a non-invasive or semi-invasive technique; calculating a ratio between at least two of the plurality of expression levels; determining whether the ratio is above or below an indicator value; and generating an indication, wherein: the indication indicates that the subject has a first condition if the ratio is above the indicator value, the indication indicates that the subject has a second condition if the ratio is below the indicator value, the first condition is different from the second condition, the first condition is atopic dermatitis or psoriasis, and the second condition is atopic dermatitis or psoriasis. 2. The method of claim 1, wherein the plurality of expression levels are determined by: a. obtaining the skin sample; b. isolating nucleic acids from the skin sample; and c. measuring the plurality of expression levels. 3. The method of claim 1 or 2, wherein the plurality of expression levels are associated with a plurality of genes. 4. The method of claim 3, wherein the plurality of genes comprises IL13, IL4R, CCL17, CCL26, IL23A, IL22, S100A7, S100A8, S100A9, CXCL9, CXCL10, CXCL 11, NOS2, IL17A, IL13R, CCL19, CCL27, TSLP, CCL18, IL23A, IL4R, and IL31. 5. The method of claim 3 or 4, wherein the plurality of genes comprises IL13, CCL17, NOS2, and IL17A. 6. The method of any one of claims 1 to 5, wherein the ratio is determined by dividing a sum of an expression level of IL13 and an expression level of CCL17 by a sum of an expression level of NOS2 and an expression level of IL17A. 7. The method of any one of claims 1 to 6, wherein: calculating the ratio between the at least two of the plurality of expression levels is performed by a machine learning model, and WSGR Docket No.44503-747.601 determining whether the ratio is above or below the indicator value is performed by the machine learning model. 8. The method of any one of claims 1 to 7, wherein the indicator value is 1. 9. The method of any one of claims 1 to 8, wherein an AUC for the indicator value is at least 0.94. 10. The method of any one of claims 1 to 9, wherein the AUC for the indicator value is at least 0.83. 11. The method of any one of claims 1 to 10, further comprising assigning a classifier to at least one expression level of the plurality of expression levels. 12. The method of claim 11, wherein generating the indication is further based on the classifier. 13. The method of any one of claims 1 to 12, wherein the non-invasive technique comprises use of one or more adhesive patches or microneedles. 14. The method of any one of claims 1 to 13, wherein use of the one or more adhesive patches collects at least 1.5 mg of skin cells. 15. The method of any one of claims 1 to 14, wherein the skin sample comprises skin cells from the stratum corneum. 16. The method of any one of claims 1 to 15, wherein the skin sample is obtained from a lesional area of the subject. 17. The method of any one of claims 1 to 16, wherein the skin sample is stable for up to 10 days. 18. The method of claim 17, wherein the skin sample is stored at 10-30 degrees C. 19. The method of any one of claims 2 to 18, wherein the isolated nucleic acids comprise DNA and/or RNA. 20. The method of any one of claims 1 to 19, wherein the skin sample comprises one or more of epithelial acanthosis and mononuclear perivascular infiltrates. 21. The method of any one of claims 2 to 20, wherein the isolated nucleic acids are amplified prior to measuring biomarkers. 22. The method of any one of claims 2 to 21, wherein the measuring comprises detecting the plurality of expression levels of one or more target genes by measuring one or more biomarkers. 23. The method of claim 22, wherein the detecting gene expression levels comprises hybridizing one or more probes to nucleic acids corresponding to the one or more target genes. WSGR Docket No.44503-747.601 24. The method of any one of claims 22 or 23, wherein the detecting gene expression levels comprises mass tagging and/or qPCR. 25. The method of any one of claims 22 to 24, wherein the one or more biomarkers comprises 2 or more target genes. 26. The method of any one of claims 22 to 25, wherein the one or more biomarkers comprises 5 or more target genes. 27. The method of any one of claims 22 to 26, wherein the one or more biomarkers comprises no more than 50 target genes. 28. The method of any one of claims 22 to 27, wherein the one or more target genes is differentially expressed in atopic dermatitis vs. psoriasis. 29. A system, comprising: one or more processors; and a memory comprising executable instructions which, when executed by the one or more processors, cause the system to: receive a plurality of expression levels, wherein the expression levels are derived from a skin sample obtained using a non-invasive or semi-invasive technique; calculate a ratio between at least two of the plurality of expression levels; determine whether the ratio is above or below an indicator value; and generate an indication, wherein: the indication indicates that the subject has a first condition if the ratio is above the indicator value, the indication indicates that the subject has a second condition if the ratio is below the indicator value, the first condition is different from the second condition, the first condition is atopic dermatitis or psoriasis, and the second condition is atopic dermatitis or psoriasis. 30. The system of claim 29, wherein the plurality of expression levels are determined by: a. obtaining the skin sample; b. isolating nucleic acids from the skin sample; and c. measuring the plurality of expression levels. 31. The system of claim 29 or 30, wherein the plurality of expression levels are associated with a plurality of genes. WSGR Docket No.44503-747.601 32. The system of claim 31, wherein the plurality of genes comprises IL13, IL4R, CCL17, CCL26, IL23A, IL22, S100A7, S100A8, S100A9, CXCL9, CXCL10, CXCL 11, NOS2, IL17A, IL13R, CCL19, CCL27, TSLP, CCL18, IL23A, IL4R, and IL31. 33. The system of claim 31 or 32, wherein the plurality of genes comprises IL13, CCL17, NOS2, and IL17A. 34. The method of any one of claims 29 to 33, wherein: calculating the ratio between the at least two of the plurality of expression levels is performed by a machine learning model, and determining whether the ratio is above or below the indicator value is performed by the machine learning model. 35. The system of any one of claims 29 to 34, wherein the ratio is determined by dividing a sum of an expression level of IL13 and an expression level of CCL17 by a sum of an expression level of NOS2 and an expression level of IL17A. 36. The system of any one of claims 29 to 35, wherein the indicator value is 1. 37. The system of any one of claims 29 to 36, wherein an AUC for the indicator value is at least 0.94. 38. The system of any one of claims 29 to 37, wherein the AUC for the indicator value is at least 0.83. 39. The system of any one of claims 29 to 38, wherein the one or more processors are further configured to cause the system to assign a classifier to at least one expression level of the plurality of expression levels. 40. The system of claim 39, wherein one or more processors being configured to cause the system to generate the indication is further based on the classifier. 41. The system of any one of claims 29 to 40, wherein the non-invasive technique comprises use of one or more adhesive patches or microneedles. 42. The system of any one of claims 29 to 41, wherein use of the one or more adhesive patches collects at least 1.5 mg of skin cells. 43. The system of any one of claims 29 to 42, wherein the skin sample comprises skin cells from the stratum corneum. 44. The system of any one of claims 29 to 43, wherein the skin sample is obtained from a lesional area of the subject. 45. The system of any one of claims 29 to 44, wherein the skin sample is stable for up to 10 days. 46. The system of claim 45, wherein the skin sample is stored at 10-30 degrees C. WSGR Docket No.44503-747.601 47. The system of any one of claims 30 to 46, wherein the isolated nucleic acids comprise DNA and/or RNA. 48. The system of any one of claims 29 to 47, wherein the skin sample comprises one or more of epithelial acanthosis and mononuclear perivascular infiltrates. 49. The system of any one of claims 30 to 48, wherein the isolated nucleic acids are amplified prior to measuring biomarkers. 50. The system of any one of claims 30 to 49, wherein the measuring comprises detecting the plurality of expression levels of one or more target genes by measuring one or more biomarkers. 51. The system of claim 50, wherein the detecting gene expression levels comprises hybridizing one or more probes to nucleic acids corresponding to the one or more target genes. 52. The system of any one of claims 50 or 51, wherein the detecting gene expression levels comprises mass tagging and/or qPCR. 53. The system of any one of claims 50 to 52, wherein the one or more biomarkers comprises 2 or more target genes. 54. The system of any one of claims 50 to 53, wherein the one or more biomarkers comprises 5 or more target genes. 55. The system of any one of claims 50 to 54, wherein the one or more biomarkers comprises no more than 50 target genes. 56. The system of any one of claims 50 to 55, wherein the one or more target genes is differentially expressed in atopic dermatitis vs. psoriasis. 57. A non-transitory, computer-readable medium comprising executable instructions, wherein a processor, when executing the executable instructions, performs a method for determining if a subject has atopic dermatitis or psoriasis, comprising: receiving a plurality of expression levels, wherein the expression levels are derived from a skin sample obtained using a non-invasive or semi-invasive technique; calculating a ratio between at least two of the plurality of expression levels; determining whether the ratio is above or below an indicator value; and generating an indication, wherein: the indication indicates that the subject has a first condition if the ratio is above the indicator value, WSGR Docket No.44503-747.601 the indication indicates that the subject has a second condition if the ratio is below the indicator value, the first condition is different from the second condition, the first condition is atopic dermatitis or psoriasis, and the second condition is atopic dermatitis or psoriasis. 58. The non-transitory, computer-readable medium of claim 57, wherein the plurality of expression levels are determined by: a. obtaining the skin sample; b. isolating nucleic acids from the skin sample; and c. measuring the plurality of expression levels. 59. The non-transitory, computer-readable medium of claim 57 or 58, wherein the plurality of expression levels are associated with a plurality of genes. 60. The non-transitory, computer-readable medium of claim 59, wherein the plurality of genes comprises IL13, IL4R, CCL17, CCL26, IL23A, IL22, S100A7, S100A8, S100A9, CXCL9, CXCL10, CXCL 11, NOS2, IL17A, IL13R, CCL19, CCL27, TSLP, CCL18, IL23Av2, IL4Rv2, and IL31. 61. The non-transitory, computer-readable medium of claim 59 or 60, wherein the plurality of genes comprises IL13, CCL17, NOS2, and IL17A. 62. The method of any one of claims 57 to 61, wherein: calculating the ratio between the at least two of the plurality of expression levels is performed by a machine learning model, and determining whether the ratio is above or below the indicator value is performed by the machine learning model. 63. The non-transitory, computer-readable medium of any one of claims 57 to 62, wherein the ratio is determined by dividing a sum of an expression level of IL13 and an expression level of CCL17 by a sum of an expression level of NOS2 and an expression level of IL17A. 64. The non-transitory, computer-readable medium of any one of claims 57 to 63, wherein the indicator value is 1. 65. The non-transitory, computer-readable medium of any one of claims 57 to 64, wherein an AUC for the indicator value is at least 0.94. 66. The non-transitory, computer-readable medium of any one of claims 57 to 65, wherein the AUC for the indicator value is at least 0.83. WSGR Docket No.44503-747.601 67. The non-transitory, computer-readable medium of any one of claims 57 to 66, further comprising assigning a classifier to at least one expression level of the plurality of expression levels. 68. The non-transitory, computer-readable medium of claim 67, wherein generating the indication is further based on the classifier. 69. The non-transitory, computer-readable medium of any one of claims 57 to 68, wherein the non-invasive technique comprises use of one or more adhesive patches or microneedles. 70. The non-transitory, computer-readable medium of any one of claims 57 to 69, wherein use of the one or more adhesive patches collects at least 1.5 mg of skin cells. 71. The non-transitory, computer-readable medium of any one of claims 57 to 70, wherein the skin sample comprises skin cells from the stratum corneum. 72. The non-transitory, computer-readable medium of any one of claims 57 to 71, wherein the skin sample is obtained from a lesional area of the subject. 73. The non-transitory, computer-readable medium of any one of claims 57 to 72, wherein the skin sample is stable for up to 10 days. 74. The non-transitory, computer-readable medium of claim 73, wherein the skin sample is stored at 10-30 degrees C. 75. The non-transitory, computer-readable medium of any one of claims 58 to 74, wherein the isolated nucleic acids comprise DNA and/or RNA. 76. The non-transitory, computer-readable medium of any one of claims 57 to 75, wherein the skin sample comprises one or more of epithelial acanthosis and mononuclear perivascular infiltrates. 77. The non-transitory, computer-readable medium of any one of claims 57 to 76, wherein the isolated nucleic acids are amplified prior to measuring biomarkers. 78. The non-transitory, computer-readable medium of any one of claims 57 to 77, wherein the measuring comprises detecting the plurality of expression levels of one or more target genes by measuring one or more biomarkers. 79. The non-transitory, computer-readable medium of claim 78, wherein the detecting gene expression levels comprises hybridizing one or more probes to nucleic acids corresponding to the one or more target genes. 80. The non-transitory, computer-readable medium of any one of claims 78 or 79, wherein the detecting gene expression levels comprises mass tagging and/or qPCR. WSGR Docket No.44503-747.601 81. The non-transitory, computer-readable medium of any one of claims 78 to 80, wherein the one or more biomarkers comprises 2 or more target genes. 82. The non-transitory, computer-readable medium of any one of claims 78 to 81, wherein the one or more biomarkers comprises 5 or more target genes. 83. The non-transitory, computer-readable medium of any one of claims 78 to 82, wherein the one or more biomarkers comprises no more than 50 target genes. 84. The non-transitory, computer-readable medium of any one of claims 78 to 83, wherein the one or more target genes is differentially expressed in atopic dermatitis vs. psoriasis. 85. A method useful for differentiating inflammatory skin diseases comprising: a. obtaining a sample, wherein the sample is obtained using a non-invasive or semi- invasive technique from a subject; b. isolating nucleic acids from the sample; c. measuring one or more biomarkers from the isolated nucleic acids; d. applying an algorithm to the one or more biomarkers to generate a pathology score; and e. identifying the sample as having atopic dermatitis or psoriasis based on the pathology score. 86. The method of claim 85, wherein the non-invasive technique comprises use of one or more adhesive patches or microneedles. 87. The method of claim 85, wherein the sample comprises skin cells from the stratum corneum or blood. 88. The method of claim 85, wherein the sample comprises one or more of epidermal keratinocytes, T cells, dendritic cells, and melanocytes. 89. The method of any one of claim 86, wherein use of the one or more adhesive patches collects at least 1.5 mg of skin cells. 90. The method of any one of claims 85-89, wherein the sample is obtained from a lesional area of the subject. WSGR Docket No.44503-747.601 91. The method of any one of claims 85-89, wherein the sample is obtained from a non- lesional area of the subject. 92. The method of any one of claims 85-91, wherein the sample is stable for up to 10 days. 93. The method of claim 92, wherein the sample is stored at 10-30 degrees C. 94. The method of any one of claims 85-93, wherein the isolated nucleic acids comprise DNA and/or RNA. 95. The method of any one of claims 85-94, wherein the sample comprises one or more of epithelial acanthosis and mononuclear perivascular infiltrates. 96. The method of any one of claims 85-95, wherein the isolated nucleic acids are amplified prior to measuring biomarkers. 97. The method of any one of claims 85-96, wherein the measuring comprises detecting gene expression levels of one or more target genes. 98. The method of claim 97, wherein the detecting gene expression levels comprises hybridizing one or more probes to nucleic acids corresponding to the one or more target genes. 99. The method of any one of claims 97-98, wherein the detecting gene expression levels comprises mass tagging and/or qPCR. 100. The method of any one of claims 85-99, wherein the one or more biomarkers comprises 2 or more target genes. 101. The method of any one of claims 85-100, wherein the one or more biomarkers comprises 5 or more target genes. 102. The method of any one of claims 85-101, wherein the one or more biomarkers comprises no more than 50 target genes. 103. The method of any one of claims 97-102, wherein the one or more target genes is differentially expressed in atopic dermatitis vs. psoriasis. WSGR Docket No.44503-747.601 104. The method of any one of claims 97-103, wherein the one or more target genes are selected from the group consisting of TNF, LINC02571, HLA-C, HCP5, LCE3A-E, PSORS1C1, IFNG, IL12B, IL1B, NFKB1A, MUC22, and IL36RN. 105. The method of any one of claims 97-104, wherein the one or more target genes are selected from the group consisting of IL4-R, FLG, SPINK5, EMSY, PBX2, FLG-AS1, TSBP1, CRCT1, STAT3, CLDN1, NLRP10, IL18R1, TNFRSF6B, TNXB, TSLP/R, and STAT6. 106. The method of any one of claims 97-105, wherein the one or more target genes are selected from the group consisting of IL-13, IL-23, IL-17A, S100A8, S100A9, CXCL9, CXCL10, CCL17 (TARC), CCL18 (PARC), CCL27 (Eotaxin-3), TLSP, and NOS2. 107. The method of any one of claims 97-106, wherein the one or more target genes are selected from the group consisting of HLA-B, KPNA3, MGMT, R3GCC1L, STEAP- AS2, PRR5L, and IL-13. 108. The method of any one of claims 97-103, wherein the one or more target genes are selected from the group consisting of IL-4R, IL-13, CCL17/TARC, CCL26/Eotaxin 3, IL- 17A, IL-22, IL-23A, NOS2, S100A8, S100A9, CXCL9, and CXCL10. 109. The method of any one of claims 97-103, wherein the one or more target genes are selected from the group consisting of IL-4R, IL-13, CCL17/TARC, and CCL26/Eotaxin 3. 110. The method of any one of claims 97-103, wherein the one or more target genes are selected from the group consisting of IL-17A, IL-22, IL-23A, NOS2, S100A8, S100A9, CXCL9, and CXCL10. 111. The method of any one of claims 97-103, wherein the one or more target genes are selected from the group consisting of IL-13, CCL17/TARC, NOS2, and IL-17A. 112. The method of any one of claims 85-111, wherein the algorithm comprises a ratio comparison of expression levels of two or more target genes. 113. The method of any one of claims 85-111, wherein the algorithm comprises a ratio comparison of expression levels of four or more target genes. WSGR Docket No.44503-747.601 114. The method of any one of claims 85-111, wherein the algorithm comprises a ratio comparison of expression levels of no more than 50 target genes. 115. The method of any one of claims 85-114, wherein the algorithm comprises a ratio comparison of CCL17 + IL-13 expression levels divided by NOS2 + IL-17A expression levels. 116. The method of any one of claims 85-115, wherein a pathology score greater than 1 is indicative of atopic dermatitis in the sample. 117. The method of any one of claims 85-115, wherein a pathology score less than 1 is indicative of psoriasis in the sample. 118. The method of any one of claims 85-117, wherein the method comprises calculating an AUC. 119. The method of claim 118, wherein the AUC is at least 0.90. 120. The method of claim 118, wherein the AUC is at least 0.94. 121. The method of any one of claims 118-120, wherein the method comprises calculating an AUC. 122. The method of claim 121, wherein the AUC is at least 0.94 and the p-value is less than 0.001. 123. The method of claim 121, wherein the AUC is at least 0.94 and the p-value is less than 0.0001. 124. The method of any one of claims 85-123, wherein the identifying the sample as having atopic dermatitis is based on a pathology score >0.9. 125. The method of any one of claims 85-123, wherein the identifying the sample as having psoriasis is based on a pathology score <0.9. 126. The method of any one of claims 85-123, wherein the identifying the sample as having atopic dermatitis is based on a pathology score >1.0. WSGR Docket No.44503-747.601 127. The method of any one of claims 85-123, wherein the identifying the sample as having psoriasis is based on a pathology score <1.0. 128. The method of any one of claims 85-127, wherein the sample is obtained from a subject suspected as having moderate to severe atopic dermatitis or moderate to severe psoriasis. 129. The method of any one of claims 85-128, wherein the pathology score further identifies a disease subtype and/or a disease endotype. 130. The method of any one of claims 85-129, wherein the method further comprises administering a treatment specific to atopic dermatitis or psoriasis based on the identification of the sample as having atopic dermatitis or psoriasis and/or the pathology score. 131. The method of any one of claims 85-130, wherein the subject is a human subject. 132. A system for differentiating inflammatory skin diseases comprising: a. a sample device configured to extract a sample from a subject; b. a sample processing device configured to extract nucleic acids and measure one or more biomarkers associated with the nucleic acids; c. at least one computer processor configured to (1) receive data comprising the one or more biomarkers and (2) execute an algorithm to process the data and output a pathology score; and d. a communication module for transmission of the pathology score or a visual interface configured to display the pathology score. 133. The system of claim 132, wherein the algorithm is configured to generate the pathology score from a ratio of one or more gene expression levels. 134. The system of any one of claims 132-133, wherein the algorithm is configured to generate the pathology score by: a. adding expression levels of a first and a second target gene to generate a first value; WSGR Docket No.44503-747.601 b. adding expression levels of a third and a fourth target gene to generate a second value; and c. dividing the first value by the second value to generate the pathology score. 135. The system of any one of claims 132-134, wherein increased expression of the first target gene and/or the second target gene are associated with atopic dermatitis. 136. The system of any one of claims 132-135, wherein increased expression of the third target gene and/or the fourth target gene are associated with psoriasis. 137. The system of any one of claims 132-136, wherein the pathology score greater than 1 is indicative of atopic dermatitis in the sample. 138. The system of any one of claims 132-136, wherein the pathology score less than 1 is indicative of psoriasis in the sample. 139. The system of any one of claims 132-138, wherein the first target gene and the second target gene are independently selected from the group consisting of IL-4R, IL-13, CCL17/TARC, and CCL26/Eotaxin 3. 140. The system of any one of claims 132-138, wherein the third target gene and the fourth target gene are independently selected from the group consisting of IL-17A, IL-22, IL-23A, NOS2, S100A8, S100A9, CXCL9, and CXCL10. 141. The system of any one of claims 132-138, wherein the first target gene and the second target gene are IL-13 and CCL17/TARC, respectively. 142. The system of any one of claims 132-138, wherein third target gene and the fourth target gene are IL-17A and NOS2, respectively. 143. The system of any one of claims 132-142, wherein the sample processing device comprises a DNA sequencer, a qPCR instrument, and/or a mass array instrument. 144. The system of any one of claims 132-143 wherein the subject is a human subject. 145. A composition comprising: WSGR Docket No.44503-747.601 a. one or more target nucleic acid molecules derived from a non-invasive sample obtained from a subject; and b. one or more probes configured to bind to the one or more target nucleic acid molecules, wherein the one or more target nucleic acid molecules correspond to one or more target genes, and wherein the one or more target genes is differentially expressed in atopic dermatitis vs. psoriasis. 146. The composition of claim 145, wherein the target nucleic acid molecules comprises DNA and/or RNA. 147. The composition of claim 146, wherein the DNA target nucleic acid molecule is a cDNA molecule. 148. The composition of any one of claims 145-147, wherein at least one of the one or more probes is hybridized to the one or more target nucleic acid molecules . 149. The composition of any one of claims 145-148, wherein the one or more target nucleic acid molecules corresponds to one or more target gene selected from the group consisting of IL-4R, IL-13, CCL17/TARC, CCL26/Eotaxin 3, IL-17A, IL-22, IL-23A, NOS2, S100A8, S100A9, CXCL9, and CXCL10. 150. The composition of any one of claims 145-148, wherein the one or more target nucleic acid molecules corresponds to one or more target gene selected from the group consisting of IL-13, CCL17/TARC, NOS2, and IL-17A. 151. The composition of any one of claims 145-150, wherein the one or more probes comprises a reporter moiety. 152. The composition of any one of claims 145-151 wherein the reporter moiety comprises a fluorescent label or a mass label. 153. The composition of any one of claims 145-152, wherein the composition further comprises at least one primer specific to the one or more target nucleic acid molecules. 154. The composition of any one of claims 145-153, wherein the composition further comprises at least one polymerase. WSGR Docket No.44503-747.601 155. The composition of any one of claims 145-154, wherein the subject is a human subject. 156. The composition of claim 155, wherein the human subject is suspected of having an inflammatory disease. 157. The composition of any one of claims 145-156, wherein the inflammatory disease is atopic dermatitis or psoriasis. 158. The composition of any one of claims 145-157, wherein the atopic dermatitis or the psoriasis is moderate to severe. 159. The composition of any one of claims 145-158, wherein the inflammatory disease is characterized by abnormalities in the skin barrier and/or chronic S. aureus colonization. 160. The composition of any one of claims 145-159, wherein the sample is an epidermal skin sample. 161. The composition of any one of claims 145-160, wherein the sample is a lesional skin sample. 162. The composition of any one of claims 145-160, wherein the sample is a non- lesional skin sample. 163. The composition of any one of claims 145-162, wherein the non-invasive sample is collected using one or more adhesive patches. 164. The composition of any one of claims 145-163, wherein the one or more adhesive patches is a DermTech SmartSticker . 165. The composition of any one of claims 145-164, wherein the differential expression of the one or more target genes provides a molecular signature indicative of atopic dermatitis or psoriasis. 166. A composition comprising: a) a buffer; b) an adhesive patch or portion of an adhesive patch; and WSGR Docket No.44503-747.601 c) an epidermal skin sample comprising one or more target molecules. 167. The composition of claim 166, wherein the one or more target molecules is selected from the group consisting of a protein, RNA, DNA, and lipid. 168. The composition of any one of claims 166-167. wherein the epidermal skin sample comprises at least 1.5 mg of stratum corneum tissue. 169. The composition of any one of claims 166-168, wherein the epidermal skin sample is a lesional skin sample. 170. The composition of any one of claims 166-168, wherein the epidermal skin sample is a non-lesional skin sample. 171. The composition of any one of claims 166-170 wherein the composition does not comprise a fixative reagent. 172. The composition of any one of claims 166-171, wherein the composition is stable for up to 10 days. 173. The composition of any one of claims 166-172, wherein the epidermal skin sample comprises one or more cell types selected from the group consisting of keratinocytes, T-cells, dendritic cells, and melanocytes. 174. The composition of any one of claims 166-173, wherein the one or more target molecules is a target nucleic acid molecule. 175. The composition of any one of claims 166-174, wherein the target nucleic acid molecule is an RNA molecule. 176. The composition of claim 175, wherein the target nucleic acid molecule is a cDNA molecule. 177. The composition of any one of claims 166-176, wherein the RNA or cDNA molecule corresponds to one or more genes selected from the group consisting of IL-4R, IL-13, CCL17/TARC, CCL26/Eotaxin 3, IL-17A, IL-22, IL-23A, NOS2, S100A8, S100A9, CXCL9, and CXCL10. WSGR Docket No.44503-747.601 178. The composition of any one of claims 166-176, wherein the RNA or cDNA molecule corresponds to one or more genes selected from the group consisting of IL-13, CCL17/TARC, IL-17A, and NOS2. 179. The composition of any one of claims 166-178, wherein the epidermal skin sample is from a region of the skin exhibiting histopathology indicative of epithelial acanthosis and/or mononuclear perivascular infiltrate. 180. The composition of any one of claims 166-179, wherein the target molecule is associated with a microbiome. 181. The composition of any one of claims 166-180, wherein the epidermal skin sample is obtained from a human subject. 182. The composition of any one of claims 166-181, wherein the human subject is suspected of having an inflammatory disease. 183. The composition of any one of claims 166-182, wherein the inflammatory disease is atopic dermatitis or psoriasis. |
WSGR Docket No.44503-747.601 compared to determining the ratio manually, which could manually require additionally resources and could take months or days. [0160] Additionally, in some embodiments, a range of ratios (also referred to as a range of “indicator values” with respect to FIGs.5-6 and 8) may be used ratio than a single value for the ratio. [0161] While the indicator value of 1.0 is described above for the genes described above, one of skill in the art would understand that the different genes may be used, and consequently, the indicator value may be a different value that 1.0. EXAMPLE SYSTEMS FOR DIAGNOSING ATOPIC DERMATITIS AND/OR PSORIASIS [0162] FIG.5 depicts an example computer system 500 for diagnosing atopic dermatitis, psoriasis, or both, in a human subject. In this depicted example, system 500 includes server 510, computing device 530, and device 560. [0163] In this depicted example, server 510 includes receiving component 512, analyzing component 516, machine learning component 518, database 520, and user interface (UI) component 522. [0164] In this depicted example, receiving component 512 further includes receptor 514. In some embodiments, server 510 may be configured to receive one or more samples (e.g., skin samples from the human subject) at receptor 514. Receptor 514 may be one or more physical devices as described above, or any combination thereof. In some embodiments, server 510 may not include receptor 514 and may instead receive data associated with the one or more samples to be analyzed by analyzing component 516. [0165] In this depicted example, analyzing component 516 may analyze the received one or more samples. For example, analyzing component may determine expression levels of one or more genes in the one or more samples. In some embodiments, the one or more genes may include IL13, IL4R, CCL17, CCL26, IL23A, IL22, S100A7, S100A8, S100A9, CXCL9, CXCL10, CXCL11, NOS2, IL17A, IL13R, CCL19, CCL27, TSLP, CCL18, IL23A, IL4R, and/or IL31 genes or combinations thereof. In some embodiments, the analyzing component 516 may receive the expression levels of the one or more genes. In some embodiments, the analyzing component 516 normalizes the expression levels of the one or more genes with respect to ACTB. [0166] In this depicted example, the machine learning component 518 may determine one or more classifiers and/or ratios based on the expression levels of the one or more genes (as described further with respect to FIG.6). In some embodiments, the machine learning WSGR Docket No.44503-747.601 component 518 may further determine indicator values based on received ratios as training data (as further described with respect to FIG.7). In some embodiments, machine learning component 518 may further adjust indicator values based on newly calculated ratios. In some embodiments, the machine learning component 518 may include one or more machine learning models. [0167] In some embodiments, machine learning component 518 may include a first machine learning model. In some embodiments, the first machine learning model may receive the expression levels of the one or more genes as input, and may determine a classifier for each of the expression levels indicating whether the subject has, is likely to have, is unlikely to have, or does not have a condition (e.g., atopic dermatitis or psoriasis). Additionally, in some embodiments, each classifier may be associated with one of the one or more genes. For example, the first machine learning model may receive expression levels for each of the IL13, IL4R, CCL17, CCL26, IL23A, IL22, S100A7, S100A8, S100A9, CXCL9, CXCL10, CXCL11, NOS2, IL17A, IL13R, CCL19, CCL27, TSLP, CCL18, IL23A, IL4R, and/or IL31 genes, and may determine a classifier to each of the genes. While certain genes are described above, these genes are exemplary, and other genes not listed may be used instead or in conjunction with the genes listed above. In some embodiments, the machine learning component 518 may normalize the one or more expression levels. In some embodiments, the analyzing component 516 may normalize the one or more expression levels before the machine learning component receives the normalized one or more expression levels. In some embodiments, analyzing component 516 may receive expression values that have already been normalized. [0168] The classifiers may be determined based on whether the expression level of the gene indicates that the subject has one of more conditions. In some embodiments, if the expression level for a particular gene falls within a first range, the expression level may indicate that the subject has a first condition (e.g., atopic dermatitis), and if the expression level for the particular gene falls within a second range, the expression level may indicate that the subject has a second condition (e.g., psoriasis). In some embodiments, the first range and the second range may overlap. In some embodiments, the first range and the second range may not overlap. Thus, in some embodiments, if the machine learning model determines that (1) the expression level falls in the first range, it may assign a first classifier to the associated gene indicating that the subject likely has atopic dermatitis, (2) the expression level falls out of the first range, it may assign a second classifier to the associated gene indicating that the subject likely does not have atopic dermatitis, (3) the expression level falls in the second range, it may assign a third classifier to the gene indicating that the subject likely has psoriasis, and (4) the expression level falls out of the WSGR Docket No.44503-747.601 second range, it may assign a fourth classifier to the gene indicating that the subject likely does not have psoriasis. The machine learning model may assign a classifier to one or more genes based on the sample. While certain classifiers are described above (e.g., indicating the subject likely has atopic dermatitis, indicating that the subject does not likely have atopic dermatitis, indicating that the subject likely has psoriasis, and indicating that the subject does not likely have psoriasis), these classifiers are exemplary, and other classifiers may be used. For example, other classifiers associated with other conditions may be used. As another example, other classifiers associated with different degrees (e.g., very likely, not very likely, extremely likely, or extremely unlikely) may be used. In some embodiments, the first machine learning model may receive feedback regarding the classifiers, and may adjust one or more parameters of the first machine learning model in response to the feedback. In some embodiments, the feedback may include one or more corrected classifiers. [0169] The machine learning component 518 may further determine one or more ratios between gene expression levels and compare the one or more ratios to one or more indicator values in order to determine if the patient has one or more of the conditions described above. In some embodiments, the one or more indicator values may be received by machine learning component 518. In some embodiments, the one or more indicator values are determined by the machine learning component 518. In some embodiments, the machine learning component 518 may determine the one or more ratios to determine if the patient has the one or more conditions using a second machine learning model. In some embodiments, the second machine learning model may receive the expression levels as described above as input, may determine the one or more ratios for the subject’s sample based on the expression levels, and may provide an indication of whether the patient has, is likely to have, is unlikely to have, or does not have the one or more conditions (as described further with respect to FIG.6) based on comparing the one or more ratios to one or more indicator values. In some embodiments, a ratio may be determined by comparing the expression level of a first gene to the expression level of a second gene from the sample. In some embodiments, a ratio may be determined by comparing the expression level of a first gene and a second gene to the expression level of a third gene and a fourth gene (as described with respect to FIG.2). In some embodiments, a ratio may be determined by comparing the expression level of a first gene to the expression level of a second gene and a third gene. In some embodiments, a ratio may be determined by comparing multiple expression levels associated with a first plurality of genes to multiple expression levels associated with a second plurality of genes. In some embodiments, a ratio may indicate that the subject has, is likely to have, is unlikely to have, or does not have atopic dermatitis. In some embodiments, a ratio may WSGR Docket No.44503-747.601 indicate that the subject has, is likely to have, is unlikely to have, or does not have psoriasis. In some embodiments, the ratio may indicate that the subject has, is likely to have, is unlikely to have, or does not have a condition by comparing the ratio determined by the machine learning model to one or more indicator values, where the one or more indicator values was determined using one or more separate sets of expression levels associated with the same genes of the ratio (e.g., by using previously determined ratio data to determine the one or more indicator values). [0170] For example, a ratio determined by using the expression levels of IL13 and CXCL9 to the expression levels of NOS2 and IL17A from the sample may be compared to an indicator value or a set of indicator values determined by comparing expression levels of IL13 and CXCL9 to NOS2 and IL17A from previous samples. For example, if the normalized expression levels for IL13, CXCL9, NOS2, and IL17A are 5.4, 7.2, 0.5, and 0.3, the ratio may be determined as 15.75 (e.g., by dividing the sum of the expression levels of IL13 and CXCL9 by the sum of expression levels for NOS2 and IL17A). The ratio may then be compared to the set of indicator values. In this particular embodiment, the set of indicator values may include values between 5.5 and 82.5 as well as values between 0.454 and 0.780. In this particular embodiment, the values between 5.5 and 82.5 indicate that a subject has or is likely to have atopic dermatitis, while the values between 0.454 and 0.780 indicate that subject has or is likely to have psoriasis. In this particular example, the ratio determined by the second machine learning model is 15.75, and thus falls in the set of indicator values between 5.5 and 82.5, indicating the subject has or is likely to have atopic dermatitis rather than psoriasis. Alternatively, in some embodiments, the ratio may be compared to a single indicator value (e.g., 12.0), and the subject may be indicated as having atopic dermatitis or psoriasis if the ratio is above or below the single indicator value. While certain values in the set of indicator values are described above, these values are exemplary and other values may be used. In some embodiments, the ranges of indicator values that indicate that a subject may have atopic dermatitis or psoriasis may overlap for specific ratios. Additionally, while the genes and ratios described above include expression values for IL13, CXCL9, NOS2, and IL17A, different genes and ratios can be used. For example, other genes that may be used in determine ratios may be only IL13 and IL17A, CXCL9 and IL17A, CLCL10 and CXCL9, or IL13R and CCL26. [0171] In some embodiments, a ratio may not fall within a set of indicator values. In those embodiments, the ratio may not indicate that the subject has or is likely to have atopic dermatitis or psoriasis. In some embodiments, a plurality of ratios may be determined. In some embodiments, one or more of the plurality of ratios may indicate that the subject has or is likely to have atopic dermatitis. In some embodiments, one or more of the plurality of ratios is likely to WSGR Docket No.44503-747.601 have psoriasis. In some embodiments, one or more ratios may indicate that the subject has or is likely to have atopic dermatitis and one or more ratios may indicate that the subject has or is likely to have psoriasis. In some embodiments, none of the ratios may indicate that the subject has or is likely to have atopic dermatitis and one or more ratios may indicate that the subject has or is likely to have psoriasis. [0172] Alternatively, in embodiments, where the ratio may be compared to a single indicator value, the ratio may be determined as either above or below the single indicator value. In some embodiments, if the ratio is determined to be above the single indicator value, the subject may be determined to have atopic dermatitis instead of psoriasis. In some embodiments, if the ratio is determined to be above the single indicator value, the subject may be determined to have psoriasis instead of atopic dermatitis. In some embodiments, if the ratio is determined to be below the single indicator value, the subject may be determined to have atopic dermatitis instead of psoriasis. In some embodiments, if the ratio is determined to be below the single indicator value, the subject may be determined to have psoriasis instead of atopic dermatitis. [0173] In some embodiments, sets of indicator values or single indicator value may be stored in database 520. In some embodiments, machine learning component 518 may retrieve the indicator values from database 520. In some embodiments, the one or more classifiers may be stored in database 520. In some embodiments, machine learning component 518 may retrieve the indicator values from database 520. [0174] Thus, the machine learning model 518 may receive the expression levels, and may provide them as input to the first machine learning model and the second machine learning model. The first machine learning model may receive the expression levels for each of the genes, and assign a classifier to each expression level, indicating whether each gene suggests that the subject has, is likely to have, is unlikely to have, or does not have at least one condition of the one or more conditions. The second machine learning model may receive the expression levels for each of the genes, and may calculate one or more ratios based on the expression levels in order to determine if the ratios fall within ranges in sets of indicator values indicating that the subject may have one or more conditions. Based on the one or more classifiers and the one or more ratios, the machine learning component 518 may determine whether the subject has the one or more conditions, and may generate an indicator 550 indicating whether the subject has the one or more conditions. [0175] The server 510 may further be configured to provide the indication 550 to computing device 530. In this depicted embodiment, computing device 530 may further include UI WSGR Docket No.44503-747.601 component 532, which may display the indication 550, thereby allowing a user of computing device 530 to view the indication. [0176] In some embodiments, the machine learning component 518 includes the first machine learning model and does not include the second machine learning model. In those embodiments, the indication 550 is generated based on the one or more classifiers for the one or more genes. In some embodiments, the machine learning component 518 includes the second machine learning model and does not include the first machine learning model. In those embodiments, the indication 550 is generated based on the one or more ratios. [0177] In some embodiments, as described above with respect to FIGs.2-4, the second machine learning model may use a single indicator value instead of a range of indicator values (e.g., the ratio of 0.9525 of FIG.2 of the ratio of 0.8364 as described with respect to FIG.4B). In those embodiments, the ratio determined based on the sample may be compared to the single indicator value. In some embodiments, if the ratio is higher than the single indicator value, the second machine learning model may provide an indication that the subject has a first condition (e.g., atopic dermatitis) instead of a second condition (e.g., psoriasis). In some embodiments, if the ratio is higher than the single indicator value, the second machine learning model may provide an indication that the subject has a second condition (e.g., psoriasis) instead of a first condition (e.g., atopic dermatitis). [0178] In some embodiments, server 510 may not include UI component 522. In some embodiments, server 510 includes UI component 522. UI component 522 may be configured to display indication 550. [0179] In some embodiments, the single indicator value or set of indicator values are determined by another machine learning model. In some embodiments, the another machine learning model is a regression model. In some embodiments, the another machine learning model may receive one or more ratios and one or more diagnoses (e.g., of whether a subject has atopic dermatitis or psoriasis based on comparing the one or more ratios to one or more indicator values) as feedback, and may adjust the single indicator value or the set of indicator values based on that feedback. [0180] FIG.6 shows an example process by machine learning component 518. In some embodiments, machine learning component 518 may include one or more machine learning models, as described with respect to FIG.5. In this depicted embodiment, machine learning component 518 has a machine learning model 602 (e.g., the first machine learning model as WSGR Docket No.44503-747.601 described with respect to FIG.5) and a machine learning model 604 (e.g., a second machine learning model as described with respect to FIG.5). [0181] In this depicted embodiment, machine learning component receives expression levels 610. Expression levels 610 includes a plurality of expression levels of genes, where the expression levels of genes are determined based on a sample taken from a subject. The genes may include one or more of the IL13, IL4R, CCL17, CCL26, IL23A, IL22, S100A7, S100A8, S100A9, CXCL9, CXCL10, CXCL11, NOS2, IL17A, IL13R, CCL19, CCL27, TSLP, CCL18, IL23A, IL4R, and/or IL31 genes. [0182] In this depicted embodiment, machine learning model 602 receives expression levels 610 and determines classifier 622, classifier 624, classifier 626, classifier 628 based on the expression levels 610 and one or more ranges of ranges 630. The machine learning model 602 may determine each of the classifiers by comparing the expression level of a gene to one or more ranges of expression levels in ranges 630 associated with the gene, where each range of ranges 630 includes a range of expression levels for the gene indicating that a subject has or is likely to have a condition. [0183] For example, a first expression level of expression levels 610 associated with a first gene may be compared to two ranges of expression levels of ranges 630 in order to determine if the first expression level falls within either of the two ranges of expression levels for the first gene. In some embodiments, the two ranges of the expression levels may overlap. Each of the two ranges may be associated with a condition, e.g., a first range of the two ranges may be associated with atopic dermatitis and a second range of the two ranges may be psoriasis. Based on if the expression level falls within each of the two ranges, the machine learning model 602 assigns a classifier indicating if the expression level of the first gene indicates that the subject has, is likely to have, is unlikely to have, or does not have the conditions associated with the one or more ranges. In this depicted example, classifier 622 indicates that the subject has or is likely to have atopic dermatitis and does not have or is not likely to have psoriasis, classifier 624 indicates that the subject has or is likely to have psoriasis and does not have or is not likely to have atopic dermatitis, classifier 626 indicates that the subject has or is likely to have both atopic dermatitis and psoriasis, and classifier 628 indicates that the subject does not have or is not likely to have either atopic dermatitis or psoriasis. While certain conditions and classifiers are listed above, these conditions and classifiers are exemplary, and other conditions and classifiers may be used. [0184] Thus, in this depicted example, if the first expression level falls within the first range of the two ranges but not the second range of the two ranges, the machine learning model 602 may assign classifier 622 to indicate that the subject has or likely has atopic dermatitis but does not WSGR Docket No.44503-747.601 have or is not likely to have psoriasis because the expression level of the first gene falls in the first range but not in the second range. If the first expression level falls out of the first range but in the second range, the machine learning model 602 may assign classifier 624 to indicate that the subject does not have or does not likely have atopic dermatitis but does have or is likely to have psoriasis because the expression level of the first gene falls out the first range but does fall in the second range. If the first expression level falls within both the first range and the second range, the machine learning model 602 may assign classifier 626 to indicate that the subject has or likely has both atopic dermatitis and psoriasis because the expression level of the first gene falls in the first range and the second range, and if the first expression level falls out of both the first range and the second range, the machine learning model 602 may assign classifier 628 to indicate that the subject does not have or does not likely have either atopic dermatitis or psoriasis because the expression level of the first gene falls out the first range and the second range. [0185] After assigning classifiers to the expression levels, the machine learning model 602 may pair a gene and the classifier associated with the gene to form a classifier gene pair of classifier gene pairs 612. Thus, for each expression level of expression levels 610, machine learning model 602 may assign a classifier indicating whether the patient has or is likely to have atopic dermatitis, psoriasis, both, or neither, and then may generate a classifier-gene pair to show what conditions the subject may have based on the expression level of the gene. [0186] In some embodiments, the machine learning model 602 does not generate a classifier gene pair, and instead only provides the classifier for each gene. [0187] In this depicted embodiment, machine learning model 604 receives expression levels 610 and determines ratio 642, ratio 644, ratio 646, and ratio 648 based on the received expression levels. The ratios 642, 644, 646, and 648 may be determined by dividing the normalized expression level of a first gene by the normalized expression level of second gene, by dividing the normalized expression level of a first gene by the sum of normalized expression levels of a second gene and a third gene, by dividing the sum of normalized expression levels of a first gene and a second gene by the sum of normalized expression levels of a third gene and a fourth gene, or by dividing the sum of normalized expression levels of a first gene and a second gene by a normalized expression ratio of a third gene. In some embodiments, a ratio may be determined by dividing the sum of normalized expression levels of four different genes by the sum of normalized expression ratios of four other genes, where the four other genes are different from the four different genes. In some embodiments, a ratio may be calculated by dividing a normalized expression level of one gene or a sum of normalized expression levels for any number of different genes by a normalized expression level of another gene or a sum of WSGR Docket No.44503-747.601 normalized expression levels for any number of other genes, wherein the other genes are not included in the different genes. [0188] After determining ratios 642, 644, 646, and 648, machine learning model 604 may compare ratios 642, 644, 646, and 648 to one or more ranges of indicator values in ranges 650 (as described above with respect to FIG.5). Each range of indicator values in ranges 650 indicates, when a ratio falls in the range of indicator values, whether a subject has a first condition (e.g., atopic dermatitis) instead of a second condition (e.g., psoriasis) or the second condition instead of the first condition. Thus, upon generating ratios 642, 644, 646, and 648, machine learning model may compare each of ratios 642, 644, 646, and 648 to an associated range of indicator values in ranges 650 to determine if the subject has or is likely to have the first condition or the second condition. In this depicted embodiment, an associated range of indicator values is associated with the same genes as the compared ratio (e.g., if a ratio is determined by dividing a first gene by a second gene, the range of indicator values is also determined using the first gene and the second gene). [0189] In some embodiments, the ranges of indicator values are determined based on previously calculated ratios from sample received from previous subjects with the first condition and/or second condition. In some embodiments, the ranges of indicator values may span from the highest ratio to the lowest ratio calculated from samples received from previous subjects. [0190] In some embodiments, only one ratio is calculated to determine if the subject has the first condition or the second condition. In some embodiments, a first ratio is calculated and determined to be inconclusive. In those embodiments, a second ratio or any number of ratios may be calculated after the first ratio is determined to be inconclusive. [0191] After calculating ratios 642, 644, 646, and 648 and comparing ratios 642, 644, 646, and 648 to associated ranges of indicator values in ranges 650, one or more ratio indications indicating whether the subject has the first condition or the second condition may be generated. [0192] Machine learning component 518 may then generate an indication 616 based on the classifier gene pairs 612 and the ratio indicators 614, where the indication 616 indicates whether the subject has, is likely to have, is unlikely to have, or does not have, the first condition, the second condition, both, or neither. In this depicted embodiment, the machine learning component 518 further provides ratio 618, which may include any number of ratios 642, 644, 646, and 648. In some embodiments, the machine learning component 518 does not provide the ratios 618. [0193] Thus, by receiving expression levels 610 machine learning component 518 may classify each gene regarding whether the subject has, is likely to have, is unlikely to have, or does not have one or more conditions, may further calculate one or more ratios regarding that the subject WSGR Docket No.44503-747.601 has, is likely to have, is unlikely to have, or does not have the one or more conditions, and may generate an indication 616 to show whether the subject has, is likely to have, is unlikely to have, or does not have the one or more conditions based on the classifications and the ratios. [0194] While four classifiers are shown for this depicted example, these classifiers are exemplary, and other classifiers may be used. For example, while the above description describes the classifiers as indicating a first condition, a second condition, both, or neither, classifiers may also only be associated with one condition or more be associated with more than two conditions. [0195] While four ratios are shown for this depicted example, these ratios are exemplary and other ratios may be used. For example, while the above description describes four ratios being calculated, only one ratio may be calculated, or any number of ratios may be calculated. [0196] As described above, in alternative embodiments, the each ratio may be compared to a respective single indicator value as opposed to a range of indicator values. In those embodiments, if the ratio is higher than the single indicator value, the machine learning model 604 may provide an indication that the subject has a first condition (e.g., atopic dermatitis) instead of a second condition (e.g., psoriasis). In some embodiments, if the ratio is higher than the single indicator value, the machine learning model 604 may provide an indication that the subject has a second condition (e.g., psoriasis) instead of a first condition (e.g., atopic dermatitis). [0197] As described above with respect to FIGs.2-5, in some embodiments, machine learning component 518 does not include machine learning model 602 (e.g., the first machine learning model as described with respect to FIGs.2-5). In those embodiments, machine learning model 604 still receives the expression levels 610, determines ratios 642, 644, 646, and 648, compares ratios 642, 644, 646, and 648 to ranges 650, and provides ratio indications. [0198] As described above with respect to FIGs.2-5 and 7, in some embodiments, a ratio based on the received sample’s gene expression levels is determined and compared to a single ratio to indicate whether or not the subject has a first condition (e.g., atopic dermatitis) instead of a second condition (e.g., psoriasis). In some embodiments, if the ratio based on the received sample is higher than the single ratio, the ratio indications 614 will indicate that the subject has the first condition instead of the second condition. In some embodiments, if the ratio based on the WSGR Docket No.44503-747.601 received sample is higher than the single ratio, the ratio indications 614 will indicate that the subject has the second condition instead of the first condition. [0199] In some embodiments, machine learning component 518 may include another machine learning model as described above, which may receive ratio data and diagnoses data as feedback to adjust one or more indicator values. [0200] FIG.7 shows an example process by machine learning component 518. In this depicted example, machine learning component 518 shows includes machine learning model 702. [0201] In this depicted embodiment, machine learning component receives expression levels 610. Expression levels 610 includes a plurality of expression levels of genes, where the expression levels of genes are determined based on a sample taken from a subject. The genes may include one or more of the IL13, IL4R, CCL17, CCL26, IL23A, IL22, S100A7, S100A8, S100A9, CXCL9, CXCL10, CXCL11, NOS2, IL17A, IL13R, CCL19, CCL27, TSLP, CCL18, IL23A, IL4R, and/or IL31 genes. As described above, these genes are exemplary, and others not listed may used instead of or in conjunction with the above genes. [0202] In this depicted embodiment, machine learning model 702 receives expression levels 610 and determines ratio 720. The ratio 720 may be determined by dividing the normalized expression level of a first gene by the normalized expression level of second gene, by dividing the normalized expression level of a first gene by the sum of normalized expression levels of a second gene and a third gene, by dividing the sum of normalized expression levels of a first gene and a second gene by the sum of normalized expression levels of a third gene and a fourth gene, or by dividing the sum of normalized expression levels of a first gene and a second gene by a normalized expression ratio of a third gene. In some embodiments, a ratio may be determined by dividing the sum of normalized expression levels of four different genes by the sum of normalized expression ratios of four other genes, where the four other genes are different from the four different genes. In some embodiments, a ratio may be calculated by dividing a normalized expression level of one gene or a sum of normalized expression levels for any number of different genes by a normalized expression level of another gene or a sum of normalized expression levels for any number of other genes, wherein the other genes are not included in the different genes. [0203] After determining ratio 720, machine learning model 702 may compare ratio 720 to indicator value 722 (as described above with respect to FIGs.2-5). If ratio 720 is higher than indicator value 722, machine learning model 702 may provide a ratio indication 714 indicating that the subject has a first condition (e.g., atopic dermatitis) rather than a second condition (e.g., psoriasis). If ratio 720 is lower than indicator value 722, machine learning model 702 may WSGR Docket No.44503-747.601 provide a ratio indication 714 indicating that the subject has the second condition rather than the first condition. [0204] Machine learning component 518 may then generate an indication 716 based on the classifier ratio indicator 714, where the indication 716 indicates whether the subject has or is likely to have the first condition instead of the second condition or the second condition instead of the first condition. In this depicted embodiment, the machine learning component 518 further provides ratio 720. [0205] Thus, by receiving expression levels 610 machine learning component 518 may calculate ratio 720 and compare it indicator value 722 to determine if the subject has or is likely to have one condition rather than another condition, and may generate an indication 716 to show whether the subject has or is likely to have the one condition or another based on the comparison. EXAMPLE SYSTEM FOR TRAINING ONE OR MORE MACHINE LEARNING MODELS [0206] FIG.8 depicts an example computer system 800 for training a machine learning model to diagnose atopic dermatitis, psoriasis, or both, in a human subject. In this depicted example, system 500 includes server 510 and computing device 830. [0207] In this depicted example, server 510 further includes receiving component 512, analyzing component 516, machine learning component 518, and database 520 as described with respect to FIG.5. [0208] In this depicted example, computing device 830 further includes UI component 832. UI component 832 may be configured to display indications received from the server 510 (e.g., indication 820) and may further be configured to receive user input (e.g., for feedback 840). In some embodiments, computing device is further configured to provide feedback 840. [0209] In this depicted embodiments, computing device is configured to provide training data 810 to server 510. Training data 810 may include one or more gene expression levels for training one or more machine learning models of machine learning component 518 (e.g., the first machine learning model as described with respect to FIG.5 or machine learning model 602 of FIG.6). The gene expression levels may further be for training one or more machine learning models (e.g., the second machine learning model as described with respect to FIG.5 or machine learning model 604 of FIG.6) to determine one or more ratios and compare ratios to one or more ranges or a single ratio (as described with respect to FIGs.2-5). In some embodiments, the training data WSGR Docket No.44503-747.601 810 may be used by another machine learning model (as described above with respect to FIGs. 5-7) to determine one or more indicator values that ratios may be compared to. [0210] Machine learning component 518 may provide an indication 820 based on the expression levels, where in the indication 820 indicates if a subject has, is likely to have, is unlikely to have, or does not have a first condition (e.g., atopic dermatitis) instead of a second condition (e.g., psoriasis), the second condition instead of the first condition, both, or neither. In some embodiments, the indication 820 may instead only indicate whether the subject has the first condition instead of the second condition after comparing the determined ratio to the single ratio. [0211] Upon receiving the indication, the computing device 830 may provide feedback 840. In some embodiments, the feedback 840 may include user input to the computing device 830 through UI component 832. In some embodiments, the feedback 840 may include an updated indication, an updated ratio, or an updated classifier. For example, if the indication 820 indicates that, based on a determined ratio, the subject has a first condition instead of a second condition, the feedback 840 may provide an updated indication showing that the subject had the second condition instead of the first condition. Based on the feedback 840, the machine learning component 518 may adjust one or more parameters of the one or more machine learning models (such as the another machine learning model, machine learning models 602 and 604 of FIG.6, or machine learning model 702 of FIG.7), thereby training the one or more machine learning models. [0212] While various embodiments of the present subject matter have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions may occur to those skilled in the art without departing from the subject matter described herein. It should be understood that various alternatives to the embodiments of the subject matter described herein may be employed. EXAMPLE METHOD FOR DIFFERENTIATING ATOPIC DERMATITIS FROM PSORIASIS [0213] FIG.9 depicts an example method 900 for determining if a subject has atopic dermatitis or psoriasis. In some embodiments, a skin sample from the subject may be analyzed to determine if the subject has atopic dermatitis or psoriasis (e.g., as described with respect to FIGs.1-2 and 5-8). In some embodiments, the skin sample is lesional. In some embodiments, the skin sample is obtained using the methods described above using a non-invasive or semi-invasive technique WSGR Docket No.44503-747.601 (e.g., through use of adhesive patches or microneedles). In some embodiments, nucleic acids are isolated from the skin sample. [0214] The method 900 begins at step 902 with receiving a plurality of expression levels, where the expression levels were derived from the skin sample that was obtained using the non-invasive or semi-invasive technique. The plurality of expression levels may be of one or more genes, where the one or more genes include any number of IL13, IL4R, CCL17, CCL26, IL23A, IL22, S100A7, S100A8, S100A9, CXCL9, CXCL10, CXCL 11,NOS2, IL17A, IL13R, CCL19, CCL27, TSLP, CCL18, IL23A, IL4R, and IL31. In some embodiments, the plurality of expression levels are normalized with respect to ACTB. [0215] At step 904, a ratio between at least two of the plurality of expression of expression levels is calculated. In some embodiments, the ratio is calculated by a machine learning model (e.g., machine learning model 604 of FIG.6 or machine learning model 702 of FIG.7). In some embodiments, the ratio is determined by dividing a sum of an expression level of IL13 and an expression level of CCL17 by a sum of an expression level of NOS2 and an expression level of IL17A. In some embodiments, at least one classifier associated with at least one expression level of the plurality of expression levels is assigned to one or more genes. In some embodiments, the classifier can indicate if the expression level indicates if the subject has atopic dermatitis, psoriasis, both, or neither. In some embodiments, another machine learning model assigns the classifier (e.g., the machine learning model 602 of FIG.6). [0216] At step 906, the ratio is compared to a second ratio. In some embodiments, the second ratio is a ratio determined based on a plurality of previous expression levels by previous patients with atopic dermatitis or psoriasis. In some embodiments, the ratio may be higher or lower than the second ratio. [0217] At step 908, an indication indicating whether the subject has atopic dermatitis or psoriasis is generated indicated based on comparing the ratio and the second ratio. In some embodiments, if the ratio is higher than the second ratio, the indication indicates that the subject has atopic dermatitis instead of psoriasis. In some embodiments, if the ratio is higher than the second ratio, the indication indicates that the subject has psoriasis instead of atopic dermatitis. In some embodiments, the indication is generated based further on the classifier. In the embodiments where the indication is generated also based on the classifier, the indication may indicate that the WSGR Docket No.44503-747.601 subject has atopic dermatitis instead of psoriasis, psoriasis instead of atopic dermatitis, has both, or has neither. [0218] Accordingly, by receiving the plurality of expression levels, calculating the ratio, comparing the ratio, and generating the indication, subjects can be diagnosed as having either atopic dermatitis or psoriasis. Computer systems [0219] The present disclosure provides computer systems that are programmed to implement methods of the disclosure. FIG.10 shows a computer system 1001 that is programmed or otherwise configured to generate indications showing if a subject has atopic dermatitis or psoriasis based on one or more calculated and/or compared ratios. The computer system 1001 can regulate various aspects of generating indications of the present disclosure, such as, for example, receiving a plurality of expression values associated with a subject, calculating one or more ratios, assigning one or more classifiers, comparing one or more ratios, or other aspects of the present disclosure. The computer system 1001 can be an electronic device of a user or a computer system that is remotely located with respect to the electronic device. The electronic device can be a mobile electronic device. [0220] The computer system 1001 includes a central processing unit (CPU, also “processor” and “computer processor” herein) 1005, which can be a single core or multi core processor, or a plurality of processors for parallel processing. The computer system 1001 also includes memory or memory location 1010 (e.g., random-access memory, read-only memory, flash memory), electronic storage unit 1015 (e.g., hard disk), communication interface 1020 (e.g., network adapter) for communicating with one or more other systems, and peripheral devices 1025, such as cache, other memory, data storage and/or electronic display adapters. The memory 1010, storage unit 1015, interface 1020 and peripheral devices 1025 are in communication with the CPU 1005 through a communication bus (solid lines), such as a motherboard. The storage unit 1015 can be a data storage unit (or data repository) for storing data. The computer system 1001 can be operatively coupled to a computer network (“network”) 1030 with the aid of the communication interface 1020. The network 1030 can be the Internet, an internet and/or extranet, or an intranet and/or extranet that is in communication with the Internet. The network 1030 in some cases is a telecommunication and/or data network. The network 1030 can include one or more computer servers, which can enable distributed computing, such as cloud computing. The network 1030, in some cases with the aid of the computer system 1001, can implement a peer-to- WSGR Docket No.44503-747.601 peer network, which may enable devices coupled to the computer system 1001 to behave as a client or a server. [0221] The CPU 1005 can execute a sequence of machine-readable instructions, which can be embodied in a program or software. The instructions may be stored in a memory location, such as the memory 1010. The instructions can be directed to the CPU 1005, which can subsequently program or otherwise configure the CPU 1005 to implement methods of the present disclosure. Examples of operations performed by the CPU 1005 can include fetch, decode, execute, and writeback. [0222] The CPU 1005 can be part of a circuit, such as an integrated circuit. One or more other components of the system 1001 can be included in the circuit. In some cases, the circuit is an application specific integrated circuit (ASIC). [0223] The storage unit 1015 can store files, such as drivers, libraries and saved programs. The storage unit 1015 can store user data, e.g., user preferences and user programs. The computer system 1001 in some cases can include one or more additional data storage units that are external to the computer system 1001, such as located on a remote server that is in communication with the computer system 1001 through an intranet or the Internet. [0224] The computer system 1001 can communicate with one or more remote computer systems through the network 1030. For instance, the computer system 1001 can communicate with a remote computer system of a user (e.g., computing device 530 of FIG.5). Examples of remote computer systems include personal computers (e.g., portable PC), slate or tablet PC’s (e.g., Apple ® iPad, Samsung ® Galaxy Tab), telephones, Smart phones (e.g., Apple ® iPhone, Android- enabled device, Blackberry ® ), or personal digital assistants. The user can access the computer system 1001 via the network 1030. [0225] Methods as described herein can be implemented by way of machine (e.g., computer processor) executable code stored on an electronic storage location of the computer system 1001, such as, for example, on the memory 1010 or electronic storage unit 1015. The machine executable or machine readable code can be provided in the form of software. During use, the code can be executed by the processor 1005. In some cases, the code can be retrieved from the storage unit 1015 and stored on the memory 1010 for ready access by the processor 1005. In some situations, the electronic storage unit 1015 can be precluded, and machine-executable instructions are stored on memory 1010. [0226] The code can be pre-compiled and configured for use with a machine having a processer adapted to execute the code, or can be compiled during runtime. The code can be supplied in a WSGR Docket No.44503-747.601 programming language that can be selected to enable the code to execute in a pre-compiled or as- compiled fashion. [0227] Aspects of the systems and methods provided herein, such as the computer system 1001, can be embodied in programming. Various aspects of the technology may be thought of as “products” or “articles of manufacture” typically in the form of machine (or processor) executable code and/or associated data that is carried on or embodied in a type of machine readable medium. Machine-executable code can be stored on an electronic storage unit, such as memory (e.g., read-only memory, random-access memory, flash memory) or a hard disk. “Storage” type media can include any or all of the tangible memory of the computers, processors or the like, or associated modules thereof, such as various semiconductor memories, tape drives, disk drives and the like, which may provide non-transitory storage at any time for the software programming. All or portions of the software may at times be communicated through the Internet or various other telecommunication networks. Such communications, for example, may enable loading of the software from one computer or processor into another, for example, from a management server or host computer into the computer platform of an application server. Thus, another type of media that may bear the software elements includes optical, electrical and electromagnetic waves, such as used across physical interfaces between local devices, through wired and optical landline networks and over various air-links. The physical elements that carry such waves, such as wired or wireless links, optical links or the like, also may be considered as media bearing the software. As used herein, unless restricted to non-transitory, tangible “storage” media, terms such as computer or machine “readable medium” refer to any medium that participates in providing instructions to a processor for execution. [0228] Hence, a machine readable medium, such as computer-executable code, may take many forms, including but not limited to, a tangible storage medium, a carrier wave medium or physical transmission medium. Non-volatile storage media include, for example, optical or magnetic disks, such as any of the storage devices in any computer(s) or the like, such as may be used to implement the databases, etc. shown in the drawings. Volatile storage media include dynamic memory, such as main memory of such a computer platform. Tangible transmission media include coaxial cables; copper wire and fiber optics, including the wires that comprise a bus within a computer system. Carrier-wave transmission media may take the form of electric or electromagnetic signals, or acoustic or light waves such as those generated during radio frequency (RF) and infrared (IR) data communications. Common forms of computer-readable media therefore include for example: a floppy disk, a flexible disk, hard disk, magnetic tape, any other magnetic medium, a CD-ROM, DVD or DVD-ROM, any other optical medium, punch WSGR Docket No.44503-747.601 cards paper tape, any other physical storage medium with patterns of holes, a RAM, a ROM, a PROM and EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave transporting data or instructions, cables or links transporting such a carrier wave, or any other medium from which a computer may read programming code and/or data. Many of these forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to a processor for execution. [0229] The computer system 1001 can include or be in communication with an electronic display 1035 that comprises a user interface (UI) 1040 for providing, for example, providing an alternate diagnosis of atopic dermatitis or psoriasis. Examples of UI’s include, without limitation, a graphical user interface (GUI) and web-based user interface. [0230] Methods and systems of the present disclosure can be implemented by way of one or more algorithms. An algorithm can be implemented by way of software upon execution by the central processing unit 1005. The algorithm can, for example, allow the system 1001 to calculate and compare ratios as well as generate indications as described herein. Web application [0231] In some embodiments, a computer program includes a web application. In light of the disclosure provided herein, those of skill in the art will recognize that a web application, in various embodiments, utilizes one or more software frameworks and one or more database systems. In some embodiments, a web application is created upon a software framework such as Microsoft ® .NET or Ruby on Rails (RoR). In some embodiments, a web application utilizes one or more database systems including, by way of non-limiting examples, relational, non-relational, object oriented, associative, XML, and document oriented database systems. In further embodiments, suitable relational database systems include, by way of non-limiting examples, Microsoft ® SQL Server, mySQL , and Oracle ® . Those of skill in the art will also recognize that a web application, in various embodiments, is written in one or more versions of one or more languages. A web application may be written in one or more markup languages, presentation definition languages, client-side scripting languages, server-side coding languages, database query languages, or combinations thereof. In some embodiments, a web application is written to some extent in a markup language such as Hypertext Markup Language (HTML), Extensible Hypertext Markup Language (XHTML), or eXtensible Markup Language (XML). In some embodiments, a web application is written to some extent in a presentation definition language such as Cascading Style Sheets (CSS). In some embodiments, a web application is written to some extent in a client-side scripting language such as Asynchronous JavaScript and XML WSGR Docket No.44503-747.601 (AJAX), Flash ® ActionScript, JavaScript, or Silverlight ® . In some embodiments, a web application is written to some extent in a server-side coding language such as Active Server Pages (ASP), ColdFusion ® , Perl, Java , JavaServer Pages (JSP), Hypertext Preprocessor (PHP), Python , Ruby, Tcl, Smalltalk, WebDNA ® , or Groovy. In some embodiments, a web application is written to some extent in a database query language such as Structured Query Language (SQL). In some embodiments, a web application integrates enterprise server products such as IBM ® Lotus Domino ® . In some embodiments, a web application includes a media player element. In various further embodiments, a media player element utilizes one or more of many suitable multimedia technologies including, by way of non-limiting examples, Adobe ® Flash ® , HTML 5, Apple ® QuickTime ® , Microsoft ® Silverlight ® , Java , and Unity ® . [0232] Referring to FIG.11, in a particular embodiment, an application provision system comprises one or more databases 1100 accessed by a relational database management system (RDBMS) 1110. Suitable RDBMSs include Firebird, MySQL, PostgreSQL, SQLite, Oracle Database, Microsoft SQL Server, IBM DB2, IBM Informix, SAP Sybase, Teradata, and the like. In this embodiment, the application provision system further comprises one or more application severs 1120 (such as Java servers, .NET servers, PHP servers, and the like) and one or more web servers 1130 (such as Apache, IIS, GWS and the like). The web server(s) optionally expose one or more web services via app application programming interfaces (APIs) 1140. Via a network, such as the Internet, the system provides browser-based and/or mobile native user interfaces. [0233] Referring to FIG.12, in a particular embodiment, an application provision system alternatively has a distributed, cloud-based architecture 1200 and comprises elastically load balanced, auto-scaling web server resources 1210 and application server resources 1220 as well synchronously replicated databases 1230. Mobile application [0234] In some embodiments, a computer program includes a mobile application provided to a mobile computing device. In some embodiments, the mobile application is provided to a mobile computing device at the time it is manufactured. In other embodiments, the mobile application is provided to a mobile computing device via the computer network described herein. [0235] In view of the disclosure provided herein, a mobile application is created by techniques known to those of skill in the art using hardware, languages, and development environments known to the art. Those of skill in the art will recognize that mobile applications are written in several languages. Suitable programming languages include, by way of non-limiting examples, WSGR Docket No.44503-747.601 C, C++, C#, Objective-C, Java , JavaScript, Pascal, Object Pascal, Python , Ruby, VB.NET, WML, and XHTML/HTML with or without CSS, or combinations thereof. [0236] Suitable mobile application development environments are available from several sources. Commercially available development environments include, by way of non-limiting examples, AirplaySDK, alcheMo, Appcelerator ® , Celsius, Bedrock, Flash Lite, .NET Compact Framework, Rhomobile, and WorkLight Mobile Platform. Other development environments are available without cost including, by way of non-limiting examples, Lazarus, MobiFlex, MoSync, and PhoneGap. Also, mobile device manufacturers distribute software developer kits including, by way of non-limiting examples, iPhone and iPad (iOS) SDK, Android SDK, BlackBerry ® SDK, BREW SDK, Palm ® OS SDK, Symbian SDK, webOS SDK, and Windows ® Mobile SDK. [0237] Those of skill in the art will recognize that several commercial forums are available for distribution of mobile applications including, by way of non-limiting examples, Apple ® App Store, Google ® Play, Chrome WebStore, BlackBerry ® App World, App Store for Palm devices, App Catalog for webOS, Windows ® Marketplace for Mobile, Ovi Store for Nokia ® devices, Samsung ® Apps, and Nintendo ® DSi Shop.
WSGR Docket No.44503-747.601 Non-transitory computer readable storage medium [0238] In some embodiments, the platforms, systems, media, and methods disclosed herein include one or more non-transitory computer readable storage media encoded with a program including instructions executable by the operating system of an optionally networked computing device. In further embodiments, a computer readable storage medium is a tangible component of a computing device. In still further embodiments, a computer readable storage medium is optionally removable from a computing device. In some embodiments, a computer readable storage medium includes, by way of non-limiting examples, CD-ROMs, DVDs, flash memory devices, solid state memory, magnetic disk drives, magnetic tape drives, optical disk drives, distributed computing systems including cloud computing systems and services, and the like. In some cases, the program and instructions are permanently, substantially permanently, semi- permanently, or non-transitorily encoded on the media. Software modules [0239] In some embodiments, the platforms, systems, media, and methods disclosed herein include software, server, and/or database modules, or use of the same. In view of the disclosure provided herein, software modules are created by techniques known to those of skill in the art using machines, software, and languages known to the art. The software modules disclosed herein are implemented in a multitude of ways. In various embodiments, a software module comprises a file, a section of code, a programming object, a programming structure, a distributed computing resource, a cloud computing resource, or combinations thereof. In further various embodiments, a software module comprises a plurality of files, a plurality of sections of code, a plurality of programming objects, a plurality of programming structures, a plurality of distributed computing resources, a plurality of cloud computing resources, or combinations thereof. In various embodiments, the one or more software modules comprise, by way of non-limiting examples, a web application, a mobile application, a standalone application, and a distributed or cloud computing application. In some embodiments, software modules are in one computer program or application. In other embodiments, software modules are in more than one computer program or application. In some embodiments, software modules are hosted on one machine. In other embodiments, software modules are hosted on more than one machine. In further embodiments, software modules are hosted on a distributed computing platform such as a cloud computing platform. In some embodiments, software modules are hosted on one or more WSGR Docket No.44503-747.601 machines in one location. In other embodiments, software modules are hosted on one or more machines in more than one location. Databases [0240] In some embodiments, the platforms, systems, media, and methods disclosed herein include one or more databases, or use of the same. In view of the disclosure provided herein, those of skill in the art will recognize that many databases are suitable for storage and retrieval of user information, gene expression levels, ratios, and classifiers. In various embodiments, suitable databases include, by way of non-limiting examples, relational databases, non-relational databases, object oriented databases, object databases, entity-relationship model databases, associative databases, XML databases, document oriented databases, and graph databases. Further non-limiting examples include SQL, PostgreSQL, MySQL, Oracle, DB2, Sybase, and MongoDB. In some embodiments, a database is Internet-based. In further embodiments, a database is web-based. In still further embodiments, a database is cloud computing-based. In a particular embodiment, a database is a distributed database. In other embodiments, a database is based on one or more local computer storage devices. EXAMPLES [0241] These examples are provided for illustrative purposes only and not to limit the scope of the claims provided herein. Example 1 – Training A Machine Learning Model to Analyze Skin Samples from Patients with AD or PS [0242] Epidermal skin samples were non-invasively collected from the lesional skin of the patients with moderate to severe AD (n=20) or moderate to severe psoriasis (n=20). RNA was isolated and analyzed by quantitative real-time PCR for the expression levels of twenty different genes. The expression levels were normalized based on ACTB being used as a housekeeping gene (e.g., to account for differences in sample input amounts). [0243] A random forest machine learning model trained to receive gene expression levels and priorities genes to be evaluated as ratios of genes elevated in AD compared to genes elevated in severe psoriasis. The best identified ratio then compared the gene expression levels of CCL17, IL13, IL17A, and NOS2 by dividing the sum of normalized expression levels of CCL17 and IL13 by the sum of normalized expression levels for IL17A and NOS 2 to generate a ratio for WSGR Docket No.44503-747.601 each skin sample to be compared to an indicator value in order to determine if each patient has atopic dermatitis or psoriasis. Example 2 – Using A Machine Learning Model to Analyze Skin Samples from Patients with AD or PS [0244] Epidermal skin samples were non-invasively collected from the lesional skin of a patient not yet diagnosed with AD or PS. RNA was isolated and analyzed by quantitative real-time PCR for the expression levels of twenty different genes. The expression levels were normalized based on ACTB being used as a housekeeping gene. [0245] The gene expression levels are provided to the machine learning model of Example 1. The machine learning model compares the gene expression levels of CCL17, IL13, IL17A, and NOS2 by dividing the sum of normalized expression levels of CCL17 and IL13 by the sum of normalized expression levels for IL17A and NOS2 to generate a ratio for the skin sample, where the ratio is then compared to an indicator value in the model. Example 3 – Further Training A Machine Learning Model to Analyze Skin Samples from Patients with AD or PS [0246] The patient of Example 2 is then diagnosed with having atopic dermatitis. The determined ratio of Example 2 is then provided to a regression machine learning model with an indicator showing that the patient was diagnosed as having atopic dermatitis. The regression machine learning model adjusts one or more parameters in response, as well as adjusting one or more associated indicator values to be provided as updated indicator values for use of the random forest model, thereby further training the machine learning model. Example 4 – Training A Machine Learning Model to Analyze Skin Samples from Patients with AD or PS [0247] Psoriasis and atopic dermatitis (AD) are two of the most prevalent chronic inflammatory skin diseases. Currently, diagnosis of psoriasis and AD is based on the combination of a skin exam and review of medical history. In some instances, the overlapping clinical characteristics and disease manifestations make it difficult to distinguish these two diseases, sometimes prompting a skin biopsy to look for the characteristic psoriatic histopathologic features. While effective, skin biopsies are invasive and have the potential for complications, especially in diseases characterized by abnormalities in the skin barrier and chronic S. aureus colonization. Here, we describe a non-invasive method to differentiate AD and psoriasis by comparing the expression of key genes involved in disease pathogenesis in AD and psoriasis. Epidermal skin samples were non-invasively collected from the lesional or nonlesional skin of the patients with moderate to severe AD (n=20) or moderate to severe psoriasis (n=20) using the DermTech Smart WSGR Docket No.44503-747.601 Sticker. RNA was isolated and analyzed by quantitative real-time PCR for the expression of IL- 13, IL-23, IL-17A, S100A8, S100A9, CXCL9, CXCL10, CCL17 (TARC), CCL18 (PARC), CCL27 (Eotaxin-3), TLSP, and NOS2. The expression levels were normalized based on ACTB being used as a housekeeping gene. [0248] A logistic regression machine learning model trained to receive gene expression levels and calculate various ratios based on the gene expression levels was provided the expression levels of the twenty different genes for each patient. [0249] Dysregulation of IL-13, CCL17, IL-17A, and NOS2 exhibited the greatest differences between psoriasis and AD. Overall, this study demonstrates the potential utility of noninvasive skin sampling to differentiate AD and psoriasis based on a molecular signature from only four genes. The ability to distinguish these two disease conditions provides a valuable asset in the hands of physicians for clinical decision-making and can be utilized for the personalized treatment of AD and psoriasis patients. Example 5 – Using A Machine Learning Model to Analyze Skin Samples from Patients with AD or PS [0250] Epidermal skin samples were non-invasively collected from the lesional skin of a patient not yet diagnosed with AD or PS. RNA was isolated and analyzed by quantitative real-time PCR for the expression levels of twenty different genes. The expression levels were normalized based on ACTB being used as a housekeeping gene. [0251] The gene expression levels are provided to the machine learning model of Example 1. The machine learning model compares the gene expression levels of IL13 and IL13R by dividing the normalized expression level of IL13 by the normalized expression level for IL13R to generate a ratio for the skin sample, where the ratio is then compared to a defined ratio in the model. The calculated ratio is determined to be above the defined ratio, indicating that the subject has psoriasis. Example 6 – Further Training A Machine Learning Model to Analyze Skin Samples from Patients with AD or PS [0252] The patient of Example 5 is then diagnosed with having atopic dermatitis. The determined ratio of example 5 is then provided to the regression machine learning model of Example 3 with an indicator showing that the patient was diagnosed as having psoriasis. The machine learning model adjusts one or more parameters in response, thereby further training the machine learning model.