Cleaning FOdb Supplementary Table 5
	Isolated the 25 low-MI features used to train ML model
		1. ABTbalance
		2. ABTdensity
		3. ABTvalence
		4. DisLength
		5. GeneralSeqDec_Best
		6. Mean_Hydropathy_FB_HSP
		7. Mean_Hydropathy_KD
		8. NLLR
		9. Number_NLS_Predicted
		10. Omega
		11. Overall_NCPR
		12. PAPAprop
		13. PPII_propensity
		14. PScoreValue
		15. PositiveResidues
		16. SeqChargeDec
		17. SeqHydDec_Best
		18. delta
		19. fraction_aromatic
		20. fraction_disorder_promoting
		21. fraction_negative
		22. fraction_polar
		23. fraction_positive
		24. fraction_proline
		25. kappa
Cleaning FOdb Supplementary Table 4
	Removed invalid FOs (puncta status = "Other" or "Nucleolar"). Remaining FOs: 178
	Total duplicated sequences: 0
	Checking for invalid characters...
		Found 1 invalid characters
		Invalid char - at index 706/706 of sequence MKRAHPEYSSSDSELDETIEVEKESADENGNLSSALGSMSPTTSSQILARKRRRGIIEKRRRDRINNSLSELRRLVPSAFEKQGSAKLEKAEILQMTVDHLKMLHTAGGKAFNNPRPGQLGRLLPNQNLPLDITLQSPTGAGPFPPIRNSSPYSVIPQPGMMGNQGMIGNQGNLGNSSTGMIGNSASRPTMPSGEWAPQSSAVRVTCAATTSAMNRPVQGGMIRNPAASIPMRPSSQPGQRQTLQSQVMNIGPSELEMNMGGPQYSQQQAPPNQTAPWPESILPIDQASFASQNRQPFGSSPDDLLCPHPAAESPSDEGALLDQLYLALRNFDGLEEIDRALGIPELVSQSQAVDPEQFSSQDSNIMLEQKAPVFPQQYASQAQMAQGSYSPMQDPNFHTMGQRPSYATLRMQPRPGLRPTGLVQNQPNQLRLQLQHRLQAQQNRQPLMNQISNVSNVNLTLRPGVPTQAPINAQMLAQRQREILNQHLRQRQMHQQQQVQQRTLMMRGQGLNMTPSMVAPSGIPATMSNPRIPQANAQQFPFPPNYGISQQPDPGFTGATTPQSPLMSPRMAHTQSPMMQQSQANPAYQAPSDINGWAQGNMGGNSMFSQQSPPHFGQQANTSMYSNNMNINVSMATNTGGMSSMNQMTGQISMTSVTSVPTSGLSSMGPEQVNDPALRGGNLFPNQLPGMDMIKQEGDTTRKYC-
	Changed FO names to Head::Tail format
	Checking for the 25 low-MI features... 25 found
		Feature ABTbalance has 12 np.nan values in the following datasets:
			Verification_Set: 12
		Feature ABTdensity has 12 np.nan values in the following datasets:
			Verification_Set: 12
		Feature ABTvalence has 12 np.nan values in the following datasets:
			Verification_Set: 12
		Feature PAPAprop has 1 np.nan values in the following datasets:
			Verification_Set: 1
		Feature PScoreValue has 1 np.nan values in the following datasets:
			Verification_Set: 1

	Puncta localization for 115 FOs where Puncta_Status==YES
		Nucleus: 	52 (45.22%)
		Cytoplasm: 	43 (37.39%)
		Both: 	20 (17.39%)
	Dataset breakdown...
		Expressed_Set: 	149 (83.71%)
			YES: 	96 (64.43%)
				Localizations...
					Nucleus: 	48 (50.00%)
					Cytoplasm: 	32 (33.33%)
					Both: 	16 (16.67%)
			NO: 	53 (35.57%)
		Verification_Set: 	29 (16.29%)
			YES: 	19 (65.52%)
				Localizations...
					Cytoplasm: 	11 (57.89%)
					Nucleus: 	4 (21.05%)
					Both: 	4 (21.05%)
			NO: 	10 (34.48%)

Making phyisochemical feature vectors.
Feature Order: 
	0. ABTbalance
	1. ABTdensity
	2. ABTvalence
	3. DisLength
	4. GeneralSeqDec_Best
	5. Mean_Hydropathy_FB_HSP
	6. Mean_Hydropathy_KD
	7. NLLR
	8. Number_NLS_Predicted
	9. Omega
	10. Overall_NCPR
	11. PAPAprop
	12. PPII_propensity
	13. PScoreValue
	14. PositiveResidues
	15. SeqChargeDec
	16. SeqHydDec_Best
	17. delta
	18. fraction_aromatic
	19. fraction_disorder_promoting
	20. fraction_negative
	21. fraction_polar
	22. fraction_positive
	23. fraction_proline
	24. kappa

Saved cleaned table S5 to cleaned_dataset_s4.csv

Saved train-test splits with nucleus, cytoplasm, and formation labels to splits.csv

Saved physicochemical embeddings as a dictionary to FOdb_physicochemical_embeddings.pkl