Skip to main content

Table 5 Classification rules identified by Logic Learning Machine applied to gene expression profiles in eight selected data sets for cancer diagnosis

From: Analyzing gene expression data for pediatric and adult cancer diagnosis using logic learning machine and standard supervised methods

OutputCondition 1Condition 2Condition 3Covering
GDS4968
 Monoc. Gamm.SNHG3_1 ≤ 9.28SNORA14B ≤ 4.3095.0%
 Monoc. Gamm.Control_3389 ≤ 8.2060.0%
 MMTHOP1 > 6.23TARP_5 ≤ 6.7185.4%
 MMC22orf23 ≤ 5.20FLJ20712 ≤ 3.1426.8%
 Smold. MMDNAJC7 > 8.13IGK_2 ≤ 10.4561DEK > 6.5097.0%
 Smold. MMHNRNPA1 > 6.4451.5%
GDS4887
 HCAQP7 ≤ 8.46100%
 Non tumorCLPX_1 > 11.4116100%
GDS4794
 Normal cellsDSCC1_1 ≤ 110.1100%
 SCLCCBX3_1 > 2232.75100%
GDS4762
 Breast cancerFMN2 < = 116.32100%
 FibroblastSHC4 > 52.20100%
GDS4471
 Classic MBEFHD2_1 > 3.87LOC100132891 ≤ 4.3788.3%
 Classic MBTCL1A > 4.6631.4%
 Other MBLOC100132891 > 4.185.47 < ZMYM5_3 ≤ 6.17 76.0%
 Other MBCHIAP2 > 3.38ZNF212 ≤ 6.4540.0%
GDS4296
 Colon cancerKLK6 > 7.71100%
 MelanomaEDNRB > 5.72279100%
 Non-SCLC5.61 < TMEM51 ≤ 6.55FAM177A1 > 8.27LINC00936 > 5.59100%
 Ovarian cancerTMEM101 ≤ 6.1585%
 Ovarian cancerMEIS1_1 > 6.7057.1%
 Renal cancerLRRN4 > 4.69APBB1IP_2 > 7.46100%
GDS3952
 Benign diseasea2.32 < IGHV7–81 ≤ 3.292.87 < BM983749 ≤ 4.06LIM2 > 4.1183.8%
 Benign diseaseaLCP2_1 > 9.07ST8SIA2_1 ≤ 2.21527.0%
 Ectopic cancersST3GAL1 > 6.55PWWP2A > 6.18100%
 Healthy controlsUSMG5 > 11.8590.3%
 Healthy controlsNUFIP2_1 > 8.8141.9%
 Breast canceraMKNK1 ≤ 3.91227762_at ≤8.3BF194770 > 2.38580.4%
 Breast cancerZNF81 ≤ 2.99MMAB_1 ≤ 4.09529.4%
 Breast cancerAU143882 > 4.5721.6%
GDS3945
 Untreated controlsCOQ10A < = 125.66100%
 Renal cancerCOQ10A > 125.66100%
  1. Monoc. Gamm. Monoclonal Gammopathy, MM Multiple Myeloma, Smold. MM Smoldering Multple Myeloma, SCLC Small Cell Lung Cancer, HC Hepatocellular Carcinoma, MB Medulloblastoma
  2. aClassification algorithm truncated to the first three rules with the highest covering