SMLG (Statistical Machine Learning Group) Discussion Forum

meninonas wrote:Professor,

Earlier today, I finished running Banjo for the PAH data. Unfortunately, neither of the runs (1, 2, 4, and 6 hours) produced three equal top BNs. I still strongly believe that the logistic regression needs to be run so the BN can be done with the relevant proteins. I think this will also help with the production of a consistent BN.

I have sent the results to Hamza. He should post at another time.

Hamza/Luis, please normalize the log scores, so we know the significance of the liklihood of the structure. There is a perl script that is posted as a public project in Git and the documentation is posted in:

Manuscripts & Documentation ‹ Implemented Modules Documentation ‹ How to run scripts/programs

Banjo 1hour run

Banjo 2 hours run

Banjo 4 hours run

Banjo, 6 hours run

The best score among the last 4 run in Banjo, is the second map of the 6 hours run with 21087.

Dr. Yoo,

Here's the results from all four runs separated by runs and then placed all together.

For 1 Hour:

Total score -21149.1337926528
Log score -21149.1338 is 99.9992652870493 % of total score
Total score -21149.1337926528
Log score -21160.9560 is 0.000733973846712013 % of total score
Total score -21149.1337926528
Log score -21167.8566 is 7.39244468246854e-07 % of total score

For 2 Hours:

Total score -21135.019289767
Log score -21135.0371 is 98.2347431779535 % of total score
Total score -21135.019289767
Log score -21139.3540 is 1.31056711136369 % of total score
Total score -21135.019289767
Log score -21140.4126 is 0.454689710766322 % of total score

For 4 Hours:

Total score -21130.5446770023
Log score -21130.5821 is 96.3268588752667 % of total score
Total score -21130.5446770023
Log score -21133.8488 is 3.67314112466307 % of total score
Total score -21130.5446770023
Log score -21158.0879 is 1.09176769177344e-10 % of total score

For 6 Hours:

Total score -21087.1786
Log score -21158.7797 is 8.01747052150605e-30 % of total score
Total score -21087.1786
Log score -21087.1786 is 100 % of total score
Total score -21087.1786
Log score -21139.1145 is 2.78307696859982e-21 % of total score

For all 12 logs scores from all 4 runs:

Total score -21087.1786
Log score -21149.1338 is 1.23936296640508e-25 % of total score
Total score -21087.1786
Log score -21160.9560 is 9.09666687363709e-31 % of total score
Total score -21087.1786
Log score -21167.8566 is 9.16198948497297e-34 % of total score
Total score -21087.1786
Log score -21135.0371 is 1.64178974196851e-19 % of total score
Total score -21087.1786
Log score -21139.3540 is 2.19034077963681e-21 % of total score
Total score -21087.1786
Log score -21140.4126 is 7.59919432539741e-22 % of total score
Total score -21087.1786
Log score -21130.5821 is 1.4128610569109e-17 % of total score
Total score -21087.1786
Log score -21133.8488 is 5.38752961756428e-19 % of total score
Total score -21087.1786
Log score -21158.0879 is 1.60133536265061e-29 % of total score
Total score -21087.1786
Log score -21158.7797 is 8.01747052150605e-30 % of total score
Total score -21087.1786
Log score -21087.1786 is 100 % of total score
Total score -21087.1786
Log score -21139.1145 is 2.78307696859982e-21 % of total score

meninonas wrote:Dr. Yoo,

Here's the results from all four runs separated by runs and then placed all together.

For 1 Hour:

Total score -21149.1337926528
Log score -21149.1338 is 99.9992652870493 % of total score
Total score -21149.1337926528
Log score -21160.9560 is 0.000733973846712013 % of total score
Total score -21149.1337926528
Log score -21167.8566 is 7.39244468246854e-07 % of total score

For 2 Hours:

Total score -21135.019289767
Log score -21135.0371 is 98.2347431779535 % of total score
Total score -21135.019289767
Log score -21139.3540 is 1.31056711136369 % of total score
Total score -21135.019289767
Log score -21140.4126 is 0.454689710766322 % of total score

For 4 Hours:

Total score -21130.5446770023
Log score -21130.5821 is 96.3268588752667 % of total score
Total score -21130.5446770023
Log score -21133.8488 is 3.67314112466307 % of total score
Total score -21130.5446770023
Log score -21158.0879 is 1.09176769177344e-10 % of total score

For 6 Hours:

Total score -21087.1786
Log score -21158.7797 is 8.01747052150605e-30 % of total score
Total score -21087.1786
Log score -21087.1786 is 100 % of total score
Total score -21087.1786
Log score -21139.1145 is 2.78307696859982e-21 % of total score

For all 12 logs scores from all 4 runs:

Total score -21087.1786
Log score -21149.1338 is 1.23936296640508e-25 % of total score
Total score -21087.1786
Log score -21160.9560 is 9.09666687363709e-31 % of total score
Total score -21087.1786
Log score -21167.8566 is 9.16198948497297e-34 % of total score
Total score -21087.1786
Log score -21135.0371 is 1.64178974196851e-19 % of total score
Total score -21087.1786
Log score -21139.3540 is 2.19034077963681e-21 % of total score
Total score -21087.1786
Log score -21140.4126 is 7.59919432539741e-22 % of total score
Total score -21087.1786
Log score -21130.5821 is 1.4128610569109e-17 % of total score
Total score -21087.1786
Log score -21133.8488 is 5.38752961756428e-19 % of total score
Total score -21087.1786
Log score -21158.0879 is 1.60133536265061e-29 % of total score
Total score -21087.1786
Log score -21158.7797 is 8.01747052150605e-30 % of total score
Total score -21087.1786
Log score -21087.1786 is 100 % of total score
Total score -21087.1786
Log score -21139.1145 is 2.78307696859982e-21 % of total score

Good job! It seems we can use only the highest scoring structure.

I applied linear and logistic regression for PAH data. Linear regression run went fine, but logistic regression had a mission output in block 1.

Hamza wrote:I applied linear and logistic regression for PAH data. Linear regression run went fine, but logistic regression had a mission output in block 1.

Please look into an option to use forward selection backward elimination option in logistic regression.