| 摘要 |
Apparatus, systems and methods for determining, for each respective phenotypic characterization in a set of {T1, . . . , Tk} characterizations, that a test specimen has the respective characterization are provided. A pairwise probability function gpq(X, Wpq), for a phenotypic pair (Tp, Tq) in {T1, . . . , Tk} is learned using a training population. Wpq is a set of parameters derived from Y for (Tp, Tq) by substituting each Y1 in Y into gpq(X, Wpq), as X, where Yi is the set of cellular constituent abundance values from sample i in the training population exhibiting Tp or Tq. The learning step is repeated for each (Tp, Tq) in {T1 . . . , Tk}, thereby deriving pairwise probability functions G={g1,2(X, W1,2), . . . , gk-1, k(X, Wk-1, k)}. Pairwise probability values P={p1,2, . . . , pk-1, k} are computed, where each ppq is equal to gpq(Z, Wpq) in G, the probability that the test specimen has Tp and not Tq, where Z is cellular constituent abundance values of the test specimen.
|
