Skip to main content

Table 2 Analysis of nodes with more than two classes, corresponding to the two areas circled in green in Fig. 6

From: Topology preserving stratification of tissue neoplasticity using Deep Neural Maps and microRNA signatures

Location

Classes

Shared miRNAs

Shared miRNAs in class-average activations

[10,0]

Neoplastic small intestine (3),

Neoplastic pancreas (1),

Neoplastic bronchus and lung (2)

miR-375 \(\uparrow\), miR-7,

miR-26a, let-7f,

let-7a \(\downarrow\)

 

[11,0]

Neoplastic small intestine (4),

Neoplastic pancreas (4),

Neoplastic bronchus and lung (3)

miR-375, miR-7,

miR-26a, let-7a,

let-7f \(\uparrow\)

 

[12,0]

Neoplastic adrenal gland (6),

Neoplastic pancreas (1),

Neoplastic paraganglion (5)

miR-7 \(\uparrow\), miR-375,

let-7a, miR-26a,

let-7f

 

[9,12]

Neoplastic connective, subcutaneous,

and other soft tissues (9),

Neoplastic skin (1),

Neoplastic breast (1)

miR-21 \(\uparrow\), miR-26a \(\downarrow\),

miR-199a-3p \(\uparrow\)

miR-125b \(\downarrow\)

  1. The first two columns show the lattice nodes and classes mapped there, with the number of samples in each class shown in brackets. The third column shows highly-activated key miRNAs shared between the sample-specific activations (within the top 75% of contributing miRNAs per class). The final column displays key miRNAs common between class-average activations, but not found within any of the sample-specific activations. Up (or down) arrows represent any miRNAs where all sample-specific activations of that miRNA are higher (or lower) than their respective class-average value