Initial metadata
----------------

Add a `gather_metadata()` function for each job. It should return a `dict` which can be nested as necessary:

{parameter: value}

or

{parameter: {subvalue1: sv1}, {subvalue2: sv2}}

then write results schema for it in `pipeliner/metadata_schema/results/`

The results schema need to describe each key value pair in the `dict`. The code below creates the schema and adds two pairs that are added to the metadata `dict` automatically by the pipeliner: `OutputFiles` and `InputFiles` . So just add entries for each key value pair in the `dict` `gather_metadata()` returned.
[http://json-schema.org/draft-04/schema#]("http://json-schema.org/draft-04/schema#")

{    
"$schema": "http://json-schema.org/draft-04/schema#",
"title": "RESULTS: relion.autopick",
"description": "CTF determination in Relion using Ctffind or GCTF",
"type": "object",
"properties": {
    "OutputFile(s)": {
        "description": "Output nodes added to the pipeline",
        "type": "array",
	    "items": {
                "type": "string"			
            }
    },
    "LogFiles": { 
	"description": "Logfiles generated by the job",
	"type": "array",
	"items": { 
		"type": "string"
	}
    },
    "YourFirstDictItem": {description, type}
}}

Here’s two jobs to get started on:

**cryoEF job type**
`pipeliner/jobs/other/model_validation_job.py`

Metadata from `cryoef_angles.log`:
{Efficiency}
{Mean PSF resolution }
{Standard deviation}
{Worst PSF resolution: {resolution, phi, Theta}}
{Best PSF resolution: {resolution, phi, Theta}}
{Distribution of PSF resolution: [ [res min, resmax, %] ]}

**Model_validation.evaluate job type**
`pipeliner/jobs/other/model_validation_job.py`

Metadata just needs the summary stats from the end of `run.out`:

Ramachandran outliers = 0.10 %
favored = 97.92 %
Rotamer outliers = 0.36 %
Peptide Plane:
Cis-proline : 8.20 %
Cis-general : 0.32 %
Twisted Proline : 0.00 %
Twisted General : 0.00 %
C-beta deviations = 0
Clashscore = 2.79 (percentile: 85.7 N=33165, 2.40A+/-0.25A)
RMS(bonds) = 0.0129
RMS(angles) = 1.46
MolProbity score = 1.09 (percentile: 95.2 N=32691, 2.40A+/-0.25A)
Resolution = 2.40
R-work = 0.3316 (percentile: 0.0 N=33089, 2.40A+/-0.25A)
Refinement program = REFMAC

Unit tests ... and we'll need to write unt tests