[WIP] Create module 3 and split into smaller pieces

[jylee-bcm]

@hyunhwan-bcm Hello, hwan.

I would like to get your review on this idea, as you asked to see the dependency between the features.

This refactoring code split the big function of getAnnotateInfoRow() into several pieces, so you can see they have limited dependency.

def f1(row):
    return getAnnotateInfoRow_3_1(row, genomeRef)

def f2(row):
    return getAnnotateInfoRow_3_2(row, moduleList, decipherSortedDf)

def f3(row):
    return getAnnotateInfoRow_3_3(row, moduleList, gnomadMetricsGeneSortedDf)

def f4(row):
    return getAnnotateInfoRow_3_4(row, moduleList, omimGeneSortedDf)

def f5(row):
    return getAnnotateInfoRow_3_5(
        row,
        clinvarGeneDf,
        clinvarAlleleDf,
        hgmdHPOScoreDf,
        moduleList,
    )

df = df.apply(f1, axis=1, result_type='expand')
df = df.apply(f2, axis=1, result_type='expand')
df = df.apply(f3, axis=1, result_type='expand')
df = df.apply(f4, axis=1, result_type='expand')
annotateInfoDf = df.apply(f5, axis=1, result_type='expand')

Please if you let me know if you find this idea can be helpful, I will finalize the PR

NOTE: This code confirmed there's no runtime error, but could not perform the output difference check yet, to show the idea of code separation quicker

[hyunhwan-bcm]

Sounds like a plan, go for it

[jylee-bcm]

I will upload a proper PR after finishing two testings.

• [x] Speed Test (whether this change degenerates the speed)
• [x] Output Integrity Test (whether the output is not changed on demo data)

[jylee-bcm]

Finished Integrity test and speed test. I will re upload a newer PR.