Table 9 Most relevant patterns used by the best three models.

From: Triclustering-based classification of longitudinal data for prognostic prediction: targeting relevant clinical endpoints in amyotrophic lateral sclerosis

C2—90 days

 Tric_18_2

[ALSFRSb=12, MITOS-stage=1]

 Tric_2_2

[ALSFRSb=12]

 Tric_11_2

[ALSFRSb=12, R=12, MITOS-stage=1]

 Tric_1_2

[ALSFRSb=12, R=12]

 Tric_2_1

[ALSFRSb=12]

C1—180 days

 Tric_232_4

[R=12, MITOS-stage=1]

 Tric_74_4

[ALSFRSb=12, R=12, MITOS-stage=1]

 Tric_109_4

[ALSFRSb=12, MITOS-stage=1]

 Tric_17_4

[ALSFRSb=12, R=12]

 Tric_19_4

[R=12]

C4—365 days

 Tric_22_1

[ALSFRSsUL=12, MITOS-stage=1]

 Tric_1_1

[ALSFRSb=12, ALSFRSsUL=12]

 Tric_19_1

[ALSFRSb=12, ALSFRSsUL=12, MITOS-stage=1]

 Tric_13_1

[ALSFRSsUL=12]

 Tric_162_1

[ALSFRSsUL=12, MITOS-stage=1]

  1. For simplicity’s sake, we reproduce here only the top 5 patterns. The terminology used is the following: each pattern’s name starts with ‘Tric’ followed by an identifier, and finally, the snapshot (bicluster) position in the set of snapshots, in which 0 is the first position.
Back to article page