Skip to main content

Table 2 Random forest classification models of season

From: The subway microbiome: seasonal dynamics and direct comparison of air and surface bacterial communities

Season

Out-of-bag estimate of error rate: 8.94%

 Confusion matrix

Autumn

Spring

Summer

Winter

Class error (%)

  Autumn

53

1

5

1

11.7

  Spring

0

54

1

1

3.6

  Summer

9

0

63

0

12.5

  Winter

2

2

0

54

6.9

Most important genera in sample classification

 Family:genera

Autumn

Spring

Summer

Winter

MDA

  Moraxellaceae:Psychrobacter

0.026

0.039

0.040

0.069

0.043

  Microbacteriaceae:Cryobacterium

0.022

0.057

0.011

0.004

0.022

  Flavobacteriaceae:Flavobacterium

0.009

0.012

0.021

0.040

0.020

  Nocardioidaceae:Nocardioides

0.020

0.024

0.011

0.013

0.016

  Flavobacteriaceae:Gillisia

0.024

0.003

0.002

0.014

0.010

  Chitinophagaceae:Ferruginibacter

0.009

0.012

0.007

0.009

0.009

  Gaiellaceae:Gaiella

0.000

0.011

0.002

0.018

0.008

  Ilumatobacteraceae:CL500-29_marine_group

0.001

0.015

0.003

0.012

0.007

  Burkholderiaceae:Polaromonas

0.008

0.015

0.001

0.002

0.006

  Rubritaleaceae:Luteolibacter

0.000

0.018

0.003

0.004

0.006

  Sphingomonadaceae:Qipengyuania

0.001

0.001

0.009

0.010

0.005

  Clostridiaceae_1:Clostridium_sensu_stricto_13

0.000

0.016

0.001

0.006

0.005

  Xanthomonadaceae:Thermomonas

0.001

0.006

0.003

0.012

0.005

  Chthoniobacteraceae:Candidatus_Udaeobacter

0.001

0.009

0.002

0.009

0.005

  Staphylococcaceae:Staphylococcus

0.012

0.007

0.002

0.000

0.005

  Microbacteriaceae:Galbitalea

0.003

0.017

0.000

0.001

0.005

  Pseudoalteromonadaceae:Pseudoalteromonas

0.003

0.000

0.010

0.004

0.005

  Phormidiaceae:Tychonema_CCAP_1459:11B

0.001

0.000

0.006

0.011

0.004

  Ilumatobacteraceae:Ilumatobacter

0.001

0.006

0.002

0.009

0.004

  Demequinaceae:Demequina

0.000

0.010

0.002

0.004

0.004

  1. Confusion matrices show the classification of samples and the associated class error. The mean decrease in model accuracy (MDA; from removing the genus in question) and mean Z-scores are given for the 20 most important genera for classifying samples