Error-domain

Go back to BCO domains.

Error Domain, acceptable range of variability “error_domain”

The error domain can be used to determine what range of input returns and outputs are within the tolerance level defined in this subdomain and therefore can be used to optimize the algorithm. It consists of two subdomains: empirical and algorithmic.

The error domain IEEE schema, and definition can be found here.

The empirical error subdomain contains empirically determined values such as limits of detectability, false positives, false negatives, statistical confidence of outcomes, etc. This can be measured by running the algorithm on multiple data samples of the usability domain or through the use of carefully designed in-silico data. For example, a set of spiked, well-characterized samples can be run through the algorithm to determine the false positives, negatives, and limits of detection.

The algorithmic subdomain is descriptive of errors that originate by the fuzziness of the algorithms, driven by stochastic processes, in dynamically parallelized multi-threaded executions, or in machine learning methodologies where the state of the machine can affect the outcome. This can be measured by taking a random subset of the data and re-running the analysis, or using some rigorous mathematical modeling of the accumulated errors and providing confidence values. For example, bootstrapping is frequently used with stochastic simulation-based algorithms to accumulate sets of outcomes and estimate statistically significant variability for the results.

For data integration, BCOs used to develop knowledge bases, the error domain can, for example, contain rules that determine inclusion in the knowledge base and reference to data that pass and fail the set of rules.

The possible keys within each subdomain are workflow-specific, a free text which should be readable for a human.

   "error_domain": {
       "empirical_error": {
           "false_negative_alignment_hits": "<0.0010", 
           "false_discovery": "<0.05"
       }, 

       "algorithmic_error": { 
           "false_positive_mutation_calls": "<0.00005", 
           "false_discovery": "0.005"
       }
   }

Error Domain Troubleshooting

When using the BioCompute Object Portal to create your own BCO or update an existing one, sometimes a bug in the error domain occurs. The portal will say there is an error when you go to publish the BCO. There is a workaround available for this bug:

1. Navigate to the Error Domain tab on the left-hand side menu.
2. You will see the editable section similar to this:

{
  "algorithmic_error": {},
  "empirical_error": {}
}

3. Carefully delete just one character, such as , or the } on the bottom, and then immediately put it back in its correct spot. This essentially is telling the portal that 'you made an edit here'.

4. Click on the next arrow. Then click 'Save BCO'.

5. Click on publish BCO, and this should fix the issue.