Feature Selection Report
Date: 2026-09-08
Analyst: Jennifer Fouquier
EXPLANA Description and Analysis Summary
EXPLANA Description
EXPLANA uses established machine-learning methods combined with innovative techniques to identify the most relevant features, from a variety of input features, that correlate with a response variable.
Input features and response variables can be numerical, categorical, or from non-normal data distributions. The relationships between selected features and the response can be linear or complex, non-linear relationships.
For longitudinal datasets, changes in features for each study identifier, often subjects, are calculated using different reference points to obtain delta datasets (First, Previous and Pairwise delta datasets). This is important because features in longitudinal studies can carry varying degrees of importance between models built using different reference points. EXPLANA automates feature selection from several models built using these datasets. This report was generated to textually and graphically summarize exploratory analysis and aid hypothesis generation.
Please ensure you understand the workflow, parameters/decisions, and that the percent variation explained (using out-of-bag (OOB) scores) is adequate for your purposes.
When you are using data without prior hypotheses, you are performing exploratory analysis and should make this clear when communicating results.
Response Variable: Crude_Rate_per_100000
Analysis Notes:
Methods
Feel free to use the following text, including citation information, for use in methods to ensure reproducibility:
EXPLANA was used for exploratory analysis to identify important
features related to the response variable,
Crude_Rate_per_100000.
A random effect of State was used to
adjust for non-independence (repeated measurements) if needed. There
were 1000 trees used per Random Forest
model with a max feature fraction of 0.2
of the input features for each split per decision tree in the forest. If
mixed effects Random Forests were needed,
20 iterations were performed. BorutaSHAP
was used to find features that perform repeatedly better than shuffled
versions of all input features. Features were considered important if
they performed better than 97% of the SHAP
importance score of the best shuffled feature using
100 trials,
p=0.05. Categorical variables were binary
encoded and
low occuring categorical values were not removed.
These methods are from an EXPLANA feature selection report (version:
2025.05.09) created on
2026-09-08. Additional information can be
found at https://github.com/JTFouquier/explana/.
Config File
analyst: Jennifer Fouquier
response_var: Crude_Rate_per_100000
include_time: 'yes'
random_effect: State
sample_id: sample_id
timepoint: Year
out: workflow-results/EXPLANA-crc-Crude_Rate_per_100000-mixed-model-2012-2022-complete-columns-20260908/
iterations: '20'
n_estimators: '1000'
max_features: '0.2'
borutashap_trials: '100'
borutashap_threshold: '97'
borutashap_p: '0.05'
analyze_original: 'yes'
analyze_first: 'yes'
analyze_previous: 'yes'
analyze_pairwise: 'yes'
absolute_values: 'no'
include_reference_values: 'no'
analysis_notes: ''
enc_percent_threshold: '0'
distance_matrices: list()
df_mod: ''
delta_df_mod: |
delta_df <- delta_df
input_datasets:
metadata:
file_path: /Users/jenniferfouquier/repos/snakemake-merf/data/crc/explana_colorectal_complete_case_columns.csv
df_mod: |
df <- df %>%
select(-Total_Count_Known) %>%
# filter(Year %in% c(2018, 2020, 2022)) %>%
select(
-N_Age_Groups_Suppressed,
-N_Age_Groups,
-Current_Smoker_Status__No,
-Alcohol_Consumption__No,
-Binge_Drinking__No,
-Diabetes__No,
-Exercise__No,
-Fair_or_Poor_Health__Fair_or_Poor_Health
)
dim_method: ''
dim_param_dict:
method: noneDirectory Links
⇨ Full Analysis Directory | ⇨ Original Dataset | ⇨ First Delta Dataset | ⇨ Previous Delta Dataset | ⇨ Pairwise Delta Dataset
Results Summary
Selected Features by Dataset
| Data | Original | First | Previous | Pairwise |
|---|---|---|---|---|
| % Variance Explained | 47.0% (46.5%) | 24.2% (26.8%) | 6.0% (11.1%) | 17.0% (17.7%) |
| N Trees | 1000 | 1000 | 1000 | 1000 |
| Feature fraction/split | 0.2 | 0.2 | 0.2 | 0.2 |
| Max Depth | 7 | 7 | 7 | 7 |
| MERF Iters. | 20 | 20 | 20 | 20 |
| BorutaSHAP Trials | 100 | 100 | 100 | 100 |
| BorutaSHAP Threshold | 97 | 97 | 97 | 97 |
| P-value | 0.05 | 0.05 | 0.05 | 0.05 |
| N Study IDs | 51 | 51 | 51 | 51 |
| N Samples | 306 | 255 | 255 | 765 |
| Input Features | 21 | 25 | 25 | 35 |
| Accepted Features | 18 | 8 | 7 | 13 |
| Tentative Features | 0 | 1 | 0 | 2 |
| Rejected Features | 3 | 16 | 18 | 20 |
| Model Type (Pass/Fail) | MERF PASS; Boruta PASS | MERF PASS; Boruta PASS | MERF PASS; Boruta PASS | MERF PASS; Boruta PASS |
Selected feature ranks from models built using Original and, for longitudinal analyses, First, Previous and Pairwise delta datasets. Selected features are shown in black and labeled with feature rank. For true/positive instances of categorical variables (indicated with “ENC” after encoding), average impact on response is shown after the rank. For numerical features, impact is not shown because the feature relationship to response can be complex, requiring further post-hoc tests or inspection of SHAP dependence plots for additional insight. Empty grey boxes indicate features included in the model for a dataset, but not selected. Long feature names may be truncated and indicated with ellipses. A comprehensive list of input features can be found in model details below.
Interpretation/Literature Search
The following links can help with hypothesis generation. Names of variables likely need modification.
Original Dataset
Model Summary
| Data | Original |
|---|---|
| Model Type (Pass/Fail) | MERF PASS; Boruta PASS |
| % Variance Explained | 47.0% (46.5%) |
| N Trees | 1000 |
| Feature fraction/split | 0.2 |
| Max Depth | 7 |
| MERF Iters. | 20 |
| BorutaSHAP Trials | 100 |
| BorutaSHAP Threshold | 97 |
| P-value | 0.05 |
| N Study IDs | 51 |
| N Samples | 306 |
| Input Features | 21 |
| Accepted Features | 18 |
| Tentative Features | 0 |
| Rejected Features | 3 |
SHAP summary beeswarm plots of feature influence on the machine learning prediction of response values. Each point represents one sample, and the horizontal position indicates impact on the response as indicated on the x-axis. Points to the left indicate a negative impact, and points to the right indicate a positive impact. The colors represent the selected feature values, where red is larger and blue is smaller. For binary encoded features (‘ENC’) red is yes[1] and blue is no[0]. SHAP is generally an improvement upon other importance scores because it provides information about both rank (how helpful the feature was compared to other features [y-axis]) and impact (a positive or negative impact on response values [x-axis]). If multiple figures are shown, scales may vary with a maximum of ten features per plot.
Selected Features
| important_features | decoded_features | feature_importance_vals |
|---|---|---|
| Diabetes__Yes | Diabetes__Yes | 0.3091 |
| BMI_Categories__Obese_BMI_30_0_99_8 | BMI_Categories__Obese_BMI_30_0_99_8 | 0.2908 |
| timepoint_explana | timepoint_explana | 0.2232 |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | BMI_Categories__Normal_Weight_BMI_18_5_24_9 | 0.1960 |
| Total_Population | Total_Population | 0.1923 |
| Overall_Health__Fair | Overall_Health__Fair | 0.1478 |
| BMI_Categories__Overweight_BMI_25_0_29_9 | BMI_Categories__Overweight_BMI_25_0_29_9 | 0.1473 |
| Overall_Health__Excellent | Overall_Health__Excellent | 0.1321 |
| Exercise__Yes | Exercise__Yes | 0.1305 |
| Current_Smoker_Status__Yes | Current_Smoker_Status__Yes | 0.1131 |
| Alcohol_Consumption__Yes | Alcohol_Consumption__Yes | 0.1093 |
| Smoker_Status__Smoke_some_days | Smoker_Status__Smoke_some_days | 0.0901 |
| Fair_or_Poor_Health__Good_or_Better_Health | Fair_or_Poor_Health__Good_or_Better_Health | 0.0859 |
| Smoker_Status__Smoke_everyday | Smoker_Status__Smoke_everyday | 0.0847 |
| Overall_Health__Very_good | Overall_Health__Very_good | 0.0728 |
| Smoker_Status__Never_smoked | Smoker_Status__Never_smoked | 0.0611 |
| Binge_Drinking__Yes | Binge_Drinking__Yes | 0.0422 |
| Overall_Health__Poor | Overall_Health__Poor | 0.0335 |
Feature Stats
| important_features | feature_importance_vals | mean | std | min | 25% | 50% | 75% | max |
|---|---|---|---|---|---|---|---|---|
| Diabetes__Yes | 0.3091 | 10.632 | 1.970 | 6.6 | 9.200 | 10.45 | 11.875 | 17.4 |
| BMI_Categories__Obese_BMI_30_0_99_8 | 0.2908 | 30.678 | 4.209 | 20.5 | 27.600 | 30.70 | 33.600 | 41.0 |
| timepoint_explana | 0.2232 | 3.500 | 1.711 | 1.0 | 2.000 | 3.50 | 5.000 | 6.0 |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | 0.1960 | 32.452 | 3.678 | 23.2 | 29.900 | 32.20 | 34.500 | 46.0 |
| Total_Population | 0.1923 | 2532625.866 | 2955131.550 | 219521.0 | 655387.750 | 1731143.50 | 2885383.500 | 16539408.0 |
| Overall_Health__Fair | 0.1478 | 12.363 | 2.162 | 8.1 | 10.700 | 12.25 | 13.875 | 17.8 |
| BMI_Categories__Overweight_BMI_25_0_29_9 | 0.1473 | 35.029 | 1.459 | 30.0 | 34.100 | 35.20 | 36.000 | 39.1 |
| Overall_Health__Excellent | 0.1321 | 18.843 | 2.860 | 12.4 | 17.025 | 18.70 | 20.500 | 28.6 |
| Exercise__Yes | 0.1305 | 76.678 | 3.824 | 67.5 | 74.400 | 76.90 | 79.400 | 84.8 |
| Current_Smoker_Status__Yes | 0.1131 | 16.892 | 3.883 | 6.7 | 14.325 | 16.60 | 19.400 | 28.3 |
| Alcohol_Consumption__Yes | 0.1093 | 53.051 | 7.694 | 29.0 | 49.100 | 53.75 | 58.800 | 67.4 |
| Smoker_Status__Smoke_some_days | 0.0901 | 4.907 | 1.007 | 2.5 | 4.125 | 4.80 | 5.600 | 8.2 |
| Fair_or_Poor_Health__Good_or_Better_Health | 0.0859 | 83.183 | 3.479 | 73.7 | 81.025 | 83.70 | 85.800 | 89.6 |
| Smoker_Status__Smoke_everyday | 0.0847 | 11.983 | 3.282 | 4.2 | 9.700 | 11.90 | 13.975 | 24.0 |
| Overall_Health__Very_good | 0.0728 | 33.186 | 2.876 | 26.3 | 30.925 | 33.65 | 35.475 | 39.3 |
| Smoker_Status__Never_smoked | 0.0611 | 58.128 | 4.960 | 46.3 | 54.900 | 57.50 | 60.800 | 75.7 |
| Binge_Drinking__Yes | 0.0422 | 16.690 | 2.982 | 9.6 | 14.725 | 16.45 | 18.200 | 25.9 |
| Overall_Health__Poor | 0.0335 | 4.456 | 1.497 | 1.8 | 3.300 | 4.20 | 5.200 | 9.4 |
Input Features
| input_features | was_selected |
|---|---|
| Total_Population | yes |
| Alcohol_Consumption__Yes | yes |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | yes |
| BMI_Categories__Obese_BMI_30_0_99_8 | yes |
| BMI_Categories__Overweight_BMI_25_0_29_9 | yes |
| Binge_Drinking__Yes | yes |
| Current_Smoker_Status__Yes | yes |
| Diabetes__No_pre_diabetes_or_borderline_diabetes | no |
| Diabetes__Yes | yes |
| Exercise__Yes | yes |
| Fair_or_Poor_Health__Good_or_Better_Health | yes |
| Overall_Health__Excellent | yes |
| Overall_Health__Fair | yes |
| Overall_Health__Good | no |
| Overall_Health__Poor | yes |
| Overall_Health__Very_good | yes |
| Smoker_Status__Former_smoker | no |
| Smoker_Status__Never_smoked | yes |
| Smoker_Status__Smoke_everyday | yes |
| Smoker_Status__Smoke_some_days | yes |
| timepoint_explana | yes |
Figures
Interpretation/Literature Search
The following links can help with hypothesis generation. Names of variables likely need modification.
Log
Binary encoded columns created for categorical input variables:BorutaSHAP Figures
Results
Model Summary
| Data | First |
|---|---|
| Model Type (Pass/Fail) | MERF PASS; Boruta PASS |
| % Variance Explained | 24.2% (26.8%) |
| N Trees | 1000 |
| Feature fraction/split | 0.2 |
| Max Depth | 7 |
| MERF Iters. | 20 |
| BorutaSHAP Trials | 100 |
| BorutaSHAP Threshold | 97 |
| P-value | 0.05 |
| N Study IDs | 51 |
| N Samples | 255 |
| Input Features | 25 |
| Accepted Features | 8 |
| Tentative Features | 1 |
| Rejected Features | 16 |
SHAP summary beeswarm plots of feature influence on the machine learning prediction of response values. Each point represents one sample, and the horizontal position indicates impact on the response as indicated on the x-axis. Points to the left indicate a negative impact, and points to the right indicate a positive impact. The colors represent the selected feature values, where red is larger and blue is smaller. For binary encoded features (‘ENC’) red is yes[1] and blue is no[0]. SHAP is generally an improvement upon other importance scores because it provides information about both rank (how helpful the feature was compared to other features [y-axis]) and impact (a positive or negative impact on response values [x-axis]). If multiple figures are shown, scales may vary with a maximum of ten features per plot.
Selected Features
| important_features | decoded_features | feature_importance_vals |
|---|---|---|
| ENC_timepoint_explana_is_1__6 | timepoint_explana | 0.2670 |
| BMI_Categories__Obese_BMI_30_0_99_8 | BMI_Categories__Obese_BMI_30_0_99_8 | 0.1807 |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | BMI_Categories__Normal_Weight_BMI_18_5_24_9 | 0.1675 |
| Smoker_Status__Smoke_everyday | Smoker_Status__Smoke_everyday | 0.1528 |
| Current_Smoker_Status__Yes | Current_Smoker_Status__Yes | 0.1382 |
| Smoker_Status__Never_smoked | Smoker_Status__Never_smoked | 0.1376 |
| Overall_Health__Excellent | Overall_Health__Excellent | 0.1360 |
| BMI_Categories__Overweight_BMI_25_0_29_9 | BMI_Categories__Overweight_BMI_25_0_29_9 | 0.0946 |
Feature Stats
| important_features | feature_importance_vals | unique | top | freq | mean | std | min | 25% | 50% | 75% | max |
|---|---|---|---|---|---|---|---|---|---|---|---|
| ENC_timepoint_explana_is_1__6 | 0.2670 | 2 | False | 204 | NA | NA | NA | NA | NA | NA | NA |
| BMI_Categories__Obese_BMI_30_0_99_8 | 0.1807 | NA | NA | 3.2792157 | 2.205558 | -1.5 | 1.50 | 2.9 | 4.9 | 11.0 | |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | 0.1675 | NA | NA | -2.4972549 | 1.746254 | -7.8 | -3.70 | -2.3 | -1.4 | 2.0 | |
| Smoker_Status__Smoke_everyday | 0.1528 | NA | NA | -2.6552941 | 1.610216 | -9.8 | -3.50 | -2.5 | -1.5 | 0.9 | |
| Current_Smoker_Status__Yes | 0.1382 | NA | NA | -3.5266667 | 2.077380 | -10.9 | -4.75 | -3.3 | -2.1 | 0.8 | |
| Smoker_Status__Never_smoked | 0.1376 | NA | NA | 3.7588235 | 2.406616 | -1.6 | 2.10 | 3.4 | 5.1 | 12.9 | |
| Overall_Health__Excellent | 0.1360 | NA | NA | 0.0168627 | 2.015489 | -5.1 | -1.40 | -0.2 | 1.2 | 6.8 | |
| BMI_Categories__Overweight_BMI_25_0_29_9 | 0.0946 | NA | NA | -0.8054902 | 1.500279 | -7.0 | -1.60 | -0.7 | 0.1 | 3.5 |
Input Features
| input_features | was_selected |
|---|---|
| Total_Population | no |
| Alcohol_Consumption__Yes | no |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | yes |
| BMI_Categories__Obese_BMI_30_0_99_8 | yes |
| BMI_Categories__Overweight_BMI_25_0_29_9 | yes |
| Binge_Drinking__Yes | no |
| Current_Smoker_Status__Yes | yes |
| Diabetes__No_pre_diabetes_or_borderline_diabetes | no |
| Diabetes__Yes | no |
| Exercise__Yes | no |
| Fair_or_Poor_Health__Good_or_Better_Health | no |
| Overall_Health__Excellent | yes |
| Overall_Health__Fair | no |
| Overall_Health__Good | no |
| Overall_Health__Poor | no |
| Overall_Health__Very_good | no |
| Smoker_Status__Former_smoker | no |
| Smoker_Status__Never_smoked | yes |
| Smoker_Status__Smoke_everyday | yes |
| Smoker_Status__Smoke_some_days | no |
| ENC_timepoint_explana_is_1__2 | no |
| ENC_timepoint_explana_is_1__3 | no |
| ENC_timepoint_explana_is_1__4 | no |
| ENC_timepoint_explana_is_1__5 | no |
| ENC_timepoint_explana_is_1__6 | yes |
Figures
Interpretation/Literature Search
The following links can help with hypothesis generation. Names of variables likely need modification.
Log
Binary encoded columns created for categorical input variables:
-> timepoint_explana: ['ENC_timepoint_explana_is_1__2', 'ENC_timepoint_explana_is_1__3', 'ENC_timepoint_explana_is_1__4', 'ENC_timepoint_explana_is_1__5', 'ENC_timepoint_explana_is_1__6']BorutaSHAP Figures
Model Summary
| Data | Previous |
|---|---|
| Model Type (Pass/Fail) | MERF PASS; Boruta PASS |
| % Variance Explained | 6.0% (11.1%) |
| N Trees | 1000 |
| Feature fraction/split | 0.2 |
| Max Depth | 7 |
| MERF Iters. | 20 |
| BorutaSHAP Trials | 100 |
| BorutaSHAP Threshold | 97 |
| P-value | 0.05 |
| N Study IDs | 51 |
| N Samples | 255 |
| Input Features | 25 |
| Accepted Features | 7 |
| Tentative Features | 0 |
| Rejected Features | 18 |
SHAP summary beeswarm plots of feature influence on the machine learning prediction of response values. Each point represents one sample, and the horizontal position indicates impact on the response as indicated on the x-axis. Points to the left indicate a negative impact, and points to the right indicate a positive impact. The colors represent the selected feature values, where red is larger and blue is smaller. For binary encoded features (‘ENC’) red is yes[1] and blue is no[0]. SHAP is generally an improvement upon other importance scores because it provides information about both rank (how helpful the feature was compared to other features [y-axis]) and impact (a positive or negative impact on response values [x-axis]). If multiple figures are shown, scales may vary with a maximum of ten features per plot.
Selected Features
| important_features | decoded_features | feature_importance_vals |
|---|---|---|
| Overall_Health__Excellent | Overall_Health__Excellent | 0.2824 |
| BMI_Categories__Overweight_BMI_25_0_29_9 | BMI_Categories__Overweight_BMI_25_0_29_9 | 0.2513 |
| Diabetes__No_pre_diabetes_or_borderline_diabetes | Diabetes__No_pre_diabetes_or_borderline_diabetes | 0.1795 |
| ENC_timepoint_explana_is_5__6 | timepoint_explana | 0.1619 |
| Overall_Health__Good | Overall_Health__Good | 0.1544 |
| Overall_Health__Fair | Overall_Health__Fair | 0.1401 |
| Fair_or_Poor_Health__Good_or_Better_Health | Fair_or_Poor_Health__Good_or_Better_Health | 0.1155 |
Feature Stats
| important_features | feature_importance_vals | unique | top | freq | mean | std | min | 25% | 50% | 75% | max |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Overall_Health__Excellent | 0.2824 | NA | NA | -0.3462745 | 3.0474683 | -6.8 | -2.0 | -0.4 | 1.05 | 6.8 | |
| BMI_Categories__Overweight_BMI_25_0_29_9 | 0.2513 | NA | NA | -0.2964706 | 1.2983133 | -4.0 | -1.1 | -0.3 | 0.50 | 3.6 | |
| Diabetes__No_pre_diabetes_or_borderline_diabetes | 0.1795 | NA | NA | 0.1525490 | 0.6003799 | -3.2 | -0.1 | 0.1 | 0.50 | 3.2 | |
| ENC_timepoint_explana_is_5__6 | 0.1619 | 2 | False | 204 | NA | NA | NA | NA | NA | NA | NA |
| Overall_Health__Good | 0.1544 | NA | NA | 0.3184314 | 2.1274392 | -5.4 | -1.1 | 0.4 | 1.60 | 7.7 | |
| Overall_Health__Fair | 0.1401 | NA | NA | 0.1654902 | 1.7743761 | -5.4 | -0.8 | 0.2 | 1.35 | 4.3 | |
| Fair_or_Poor_Health__Good_or_Better_Health | 0.1155 | NA | NA | -0.0447059 | 2.4394319 | -5.3 | -1.6 | -0.2 | 1.20 | 6.6 |
Input Features
| input_features | was_selected |
|---|---|
| Total_Population | no |
| Alcohol_Consumption__Yes | no |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | no |
| BMI_Categories__Obese_BMI_30_0_99_8 | no |
| BMI_Categories__Overweight_BMI_25_0_29_9 | yes |
| Binge_Drinking__Yes | no |
| Current_Smoker_Status__Yes | no |
| Diabetes__No_pre_diabetes_or_borderline_diabetes | yes |
| Diabetes__Yes | no |
| Exercise__Yes | no |
| Fair_or_Poor_Health__Good_or_Better_Health | yes |
| Overall_Health__Excellent | yes |
| Overall_Health__Fair | yes |
| Overall_Health__Good | yes |
| Overall_Health__Poor | no |
| Overall_Health__Very_good | no |
| Smoker_Status__Former_smoker | no |
| Smoker_Status__Never_smoked | no |
| Smoker_Status__Smoke_everyday | no |
| Smoker_Status__Smoke_some_days | no |
| ENC_timepoint_explana_is_1__2 | no |
| ENC_timepoint_explana_is_2__3 | no |
| ENC_timepoint_explana_is_3__4 | no |
| ENC_timepoint_explana_is_4__5 | no |
| ENC_timepoint_explana_is_5__6 | yes |
Figures
Interpretation/Literature Search
The following links can help with hypothesis generation. Names of variables likely need modification.
| important_features | url |
|---|---|
| Overall_Health__Excellent | https://pubmed.ncbi.nlm.nih.gov/?term=Overall_Health__Excellent%20AND%20Crude_Rate_per_100000 |
| BMI_Categories__Overweight_BMI_25_0_29_9 | https://pubmed.ncbi.nlm.nih.gov/?term=BMI_Categories__Overweight_BMI_25_0_29_9%20AND%20Crude_Rate_per_100000 |
| Diabetes__No_pre_diabetes_or_borderline_diabetes | https://pubmed.ncbi.nlm.nih.gov/?term=Diabetes__No_pre_diabetes_or_borderline_diabetes%20AND%20Crude_Rate_per_100000 |
| ENC_timepoint_explana_is_5__6 | https://pubmed.ncbi.nlm.nih.gov/?term=5%20AND%206%20AND%20timepoint_explana%20AND%20Crude_Rate_per_100000 |
| Overall_Health__Good | https://pubmed.ncbi.nlm.nih.gov/?term=Overall_Health__Good%20AND%20Crude_Rate_per_100000 |
| Overall_Health__Fair | https://pubmed.ncbi.nlm.nih.gov/?term=Overall_Health__Fair%20AND%20Crude_Rate_per_100000 |
| Fair_or_Poor_Health__Good_or_Better_Health | https://pubmed.ncbi.nlm.nih.gov/?term=Fair_or_Poor_Health__Good_or_Better_Health%20AND%20Crude_Rate_per_100000 |
Log
Binary encoded columns created for categorical input variables:
-> timepoint_explana: ['ENC_timepoint_explana_is_1__2', 'ENC_timepoint_explana_is_2__3', 'ENC_timepoint_explana_is_3__4', 'ENC_timepoint_explana_is_4__5', 'ENC_timepoint_explana_is_5__6']BorutaSHAP Figures
Model Summary
| Data | Pairwise |
|---|---|
| Model Type (Pass/Fail) | MERF PASS; Boruta PASS |
| % Variance Explained | 17.0% (17.7%) |
| N Trees | 1000 |
| Feature fraction/split | 0.2 |
| Max Depth | 7 |
| MERF Iters. | 20 |
| BorutaSHAP Trials | 100 |
| BorutaSHAP Threshold | 97 |
| P-value | 0.05 |
| N Study IDs | 51 |
| N Samples | 765 |
| Input Features | 35 |
| Accepted Features | 13 |
| Tentative Features | 2 |
| Rejected Features | 20 |
SHAP summary beeswarm plots of feature influence on the machine learning prediction of response values. Each point represents one sample, and the horizontal position indicates impact on the response as indicated on the x-axis. Points to the left indicate a negative impact, and points to the right indicate a positive impact. The colors represent the selected feature values, where red is larger and blue is smaller. For binary encoded features (‘ENC’) red is yes[1] and blue is no[0]. SHAP is generally an improvement upon other importance scores because it provides information about both rank (how helpful the feature was compared to other features [y-axis]) and impact (a positive or negative impact on response values [x-axis]). If multiple figures are shown, scales may vary with a maximum of ten features per plot.
Selected Features
| important_features | decoded_features | feature_importance_vals |
|---|---|---|
| Overall_Health__Excellent | Overall_Health__Excellent | 0.2424 |
| BMI_Categories__Overweight_BMI_25_0_29_9 | BMI_Categories__Overweight_BMI_25_0_29_9 | 0.2281 |
| BMI_Categories__Obese_BMI_30_0_99_8 | BMI_Categories__Obese_BMI_30_0_99_8 | 0.2234 |
| Current_Smoker_Status__Yes | Current_Smoker_Status__Yes | 0.1958 |
| Smoker_Status__Smoke_everyday | Smoker_Status__Smoke_everyday | 0.1812 |
| Smoker_Status__Never_smoked | Smoker_Status__Never_smoked | 0.0930 |
| Overall_Health__Good | Overall_Health__Good | 0.0920 |
| Fair_or_Poor_Health__Good_or_Better_Health | Fair_or_Poor_Health__Good_or_Better_Health | 0.0639 |
| ENC_timepoint_explana_is_1__6 | timepoint_explana | 0.0607 |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | BMI_Categories__Normal_Weight_BMI_18_5_24_9 | 0.0557 |
| Overall_Health__Fair | Overall_Health__Fair | 0.0552 |
| Overall_Health__Poor | Overall_Health__Poor | 0.0387 |
| Total_Population | Total_Population | 0.0275 |
Feature Stats
| important_features | feature_importance_vals | unique | top | freq | mean | std | min | 25% | 50% | 75% | max |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Overall_Health__Excellent | 0.2424 | NA | NA | -0.1028758 | 2.617249e+00 | -6.8 | -1.8 | -0.4 | 1.5 | 7.5 | |
| BMI_Categories__Overweight_BMI_25_0_29_9 | 0.2281 | NA | NA | -0.6292810 | 1.412326e+00 | -7.0 | -1.5 | -0.6 | 0.3 | 3.6 | |
| BMI_Categories__Obese_BMI_30_0_99_8 | 0.2234 | NA | NA | 2.6028758 | 2.071015e+00 | -2.3 | 1.1 | 2.4 | 3.9 | 11.0 | |
| Current_Smoker_Status__Yes | 0.1958 | NA | NA | -2.7436601 | 1.878823e+00 | -10.9 | -3.8 | -2.5 | -1.4 | 1.5 | |
| Smoker_Status__Smoke_everyday | 0.1812 | NA | NA | -1.9368627 | 1.498708e+00 | -9.8 | -2.8 | -1.8 | -0.9 | 1.6 | |
| Smoker_Status__Never_smoked | 0.0930 | NA | NA | 2.8942484 | 2.203547e+00 | -3.4 | 1.3 | 2.6 | 4.1 | 12.9 | |
| Overall_Health__Good | 0.0920 | NA | NA | 0.3162092 | 1.967258e+00 | -6.9 | -1.0 | 0.4 | 1.6 | 7.7 | |
| Fair_or_Poor_Health__Good_or_Better_Health | 0.0639 | NA | NA | 0.4349020 | 2.131108e+00 | -5.3 | -0.9 | 0.2 | 1.8 | 6.9 | |
| ENC_timepoint_explana_is_1__6 | 0.0607 | 2 | False | 714 | NA | NA | NA | NA | NA | NA | NA |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | 0.0557 | NA | NA | -1.9955556 | 1.672751e+00 | -7.8 | -3.0 | -1.9 | -0.9 | 4.2 | |
| Overall_Health__Fair | 0.0552 | NA | NA | 0.0033987 | 1.581723e+00 | -5.4 | -1.0 | 0.1 | 1.1 | 4.3 | |
| Overall_Health__Poor | 0.0387 | NA | NA | -0.4487582 | 8.828732e-01 | -4.0 | -1.0 | -0.4 | 0.1 | 2.4 | |
| Total_Population | 0.0275 | NA | NA | 34195.7372549 | 1.487391e+05 | -513755.0 | -8918.0 | 4751.0 | 39142.0 | 1541102.0 |
Input Features
| input_features | was_selected |
|---|---|
| Total_Population | yes |
| Alcohol_Consumption__Yes | no |
| BMI_Categories__Normal_Weight_BMI_18_5_24_9 | yes |
| BMI_Categories__Obese_BMI_30_0_99_8 | yes |
| BMI_Categories__Overweight_BMI_25_0_29_9 | yes |
| Binge_Drinking__Yes | no |
| Current_Smoker_Status__Yes | yes |
| Diabetes__No_pre_diabetes_or_borderline_diabetes | no |
| Diabetes__Yes | no |
| Exercise__Yes | no |
| Fair_or_Poor_Health__Good_or_Better_Health | yes |
| Overall_Health__Excellent | yes |
| Overall_Health__Fair | yes |
| Overall_Health__Good | yes |
| Overall_Health__Poor | yes |
| Overall_Health__Very_good | no |
| Smoker_Status__Former_smoker | no |
| Smoker_Status__Never_smoked | yes |
| Smoker_Status__Smoke_everyday | yes |
| Smoker_Status__Smoke_some_days | no |
| ENC_timepoint_explana_is_1__2 | no |
| ENC_timepoint_explana_is_1__3 | no |
| ENC_timepoint_explana_is_1__4 | no |
| ENC_timepoint_explana_is_1__5 | no |
| ENC_timepoint_explana_is_1__6 | yes |
| ENC_timepoint_explana_is_2__3 | no |
| ENC_timepoint_explana_is_2__4 | no |
| ENC_timepoint_explana_is_2__5 | no |
| ENC_timepoint_explana_is_2__6 | no |
| ENC_timepoint_explana_is_3__4 | no |
| ENC_timepoint_explana_is_3__5 | no |
| ENC_timepoint_explana_is_3__6 | no |
| ENC_timepoint_explana_is_4__5 | no |
| ENC_timepoint_explana_is_4__6 | no |
| ENC_timepoint_explana_is_5__6 | no |
Figures
Interpretation/Literature Search
The following links can help with hypothesis generation. Names of variables likely need modification.
Log
Binary encoded columns created for categorical input variables:
-> timepoint_explana: ['ENC_timepoint_explana_is_1__2', 'ENC_timepoint_explana_is_1__3', 'ENC_timepoint_explana_is_1__4', 'ENC_timepoint_explana_is_1__5', 'ENC_timepoint_explana_is_1__6', 'ENC_timepoint_explana_is_2__3', 'ENC_timepoint_explana_is_2__4', 'ENC_timepoint_explana_is_2__5', 'ENC_timepoint_explana_is_2__6', 'ENC_timepoint_explana_is_3__4', 'ENC_timepoint_explana_is_3__5', 'ENC_timepoint_explana_is_3__6', 'ENC_timepoint_explana_is_4__5', 'ENC_timepoint_explana_is_4__6', 'ENC_timepoint_explana_is_5__6']