In find_transform_matrix:
	Start with the base coordinates (ALA.pdb...taken from ALA 360 of mbp).
		Use UUU for 359 (since weird problems with sidechain building crop up if there is only one residue)
	Use ALIGN_STRUCTURES to align backbone atoms to each relevant position of the SH3-bound peptide in turn
	Use the resulting transform matricies as ligamers (in align_X.pdb files)
		align_1 = 73
		align_2 = 74
		75 is central PRO; kept fixed and in the rigid prt 
		align_3 = 76
		align_4 = 77	
	Also measure the N->CA->C->O dihedrals in rasmol
	Summarized in transform_matricies

In generate_sidechain_base_coords/protein_sidechains/, build each sidechain off the ALA.pdb backbone:
	create ALA_bkbn.pdb; ALA 360 of mbp. 
	replace ALA residuetype in ALA_bkbn.pdb w/ residuetype of interest
	run generate_sidechain_base_coords.input (an HBUILD job)
		this will build the coordinates off the the backbone, with all chi=180
	repeat for all residuetypes

Copy and modify
Swap the N and CA. 
Get rid of HA (now the amide proton) since the CB is coming off the N
Rename the residues.
Rename backbone atoms:
	CA = CA1
	N = N1
	C = C1
	O = O1
Rename H (amide proton) to H1A (CA proton)
Insert new CA proton
ATOM      7  H2A ALA   360      30.022  26.847  41.253  0.00  0.00

A	ALA	ala		a	31
D	ASP	asp		J	32
E	GLU	glu		j	33
F	PHE	phe		f	34
H	HIS	his		o	35
I	ILE	ile		i	36
K	LYS	lys		B	37
L	LEU	leu		l	38
M	MET	met		m	39
N	ASN	asn		n	40
Q	GLN	gln		q	41
R	ARG	arg		r	42
S	SER	ser		s	43
T	THR	thr		t	44		
V	VAL	val		v	45
W	TRP	trp		w	46	
Y	TYR	tyr		b	47
		prt		p	48	(pro and ter)

in "make_ligand_sidechain_script":
cat protein_sidechains/ILE.pdb new_HA | grep ATOM | awk '{sub("ILE","ile"); print;}' | awk '{sub(" CA "," XX "); print;}' | awk '{sub(" N "," CA"); print;}' | awk '{sub("XX","N "); print}' | awk '{sub("CA ","CA1"); print}' | awk '{sub("N ","N1"); print}' | awk '{sub("C ","C1"); print}'  | awk '{sub("O ","O1"); print}'  | awk '{sub("HA ","XXX"); print}' | grep -v XXX | awk '{sub("H  ","H1A"); print;}' | awk '{sub("ALA","ile"); print;}' | sort -k2 -n > ligand_sidechains/ala.pdb

in "make_ligand_param_script":
../../../../../../energy_function/ligands/ExtendedSidechain ../../../../../../energy_function/ligands/AUU.pdb ligand_sidechains/ala.pdb -reslib_file  ../../../../../../energy_function/ligands/old_resparam_file_for_ligand_building -anchor_res 216 -anchor_chain Q -res_outfile ligand_params/ala.param -remove_flexibility true

C1 is picked as BAD by ExtendedSidechain; need to define the atomtype
	C1 = atomtype CN
	CA1 = atomtype CT not CN		
Need to set the virtual backbone atomtypes to U
Ligand backbone O1 should be atomtype O not OS
H1A and H2A should be HC

In "modify_ligand_params_script":
cat ligand_params/ile.param | awk '{sub("CA1  CN","CA1  CT"); print;}' | awk '{sub("C1   BAD","C1   CN "); print;}' | awk '{sub("N     N","N     U"); print;}' | awk '{sub("H     HN","H     U "); print;}' | awk '{sub("CA    CT","CA    U "); print;}' | awk '{sub("HA    CH","HA    U "); print;}' | awk '{sub("CB    CH3","CB    U  "); print;}' | awk '{sub("C     CN","C     U "); print;}' | awk '{sub("O     O","O     U"); print;}' | awk '{sub("O1   OS","O1   O "); print;}' | awk '{sub("A  HD","A  HC"); print;}' > t1; mv -f t1 ligand_params/ile.param 

Set charges (use the charges from the main resparam file)
	Modifications (based on OPLS-AA rules and PRO charges):
		For CA, use -0.05
		For CB (now coming off the N), use the resparam charge + 0.07

Set rings and other rigid groups of atoms to the same other_info number
Define rotatable bonds with "-" sign in dihedral columns


only one CB allowed (one is already there as a virtual bkbn atom)....rename ligand CB to CB1

In "assemble_ligand_file_script":
cat ligand_file_top ligand_params/ala.param ligand_file_middle ligand_sidechains/ala.pdb ligand_file_end | awk '{sub("xxx","ala"); print;}' | awk '{sub("x","a"); print;}' | awk '{sub("rescode","31"); print;}' | awk '{sub("num_chi","13"); print;}' | awk '{sub("17   CB ","17   CB1"); print;}' | awk '{sub("ATOM      5  CB ","ATOM      5  CB1"); print;}' > ligand_files/ala.ligand

Ligamers:
There are four positions in the ligand.
For each, there is a transform matrix.
Additional entries are for rotatable bonds; the last is the seq_position text for that ligamer
	the first rotable bond is the CA1->N->C-O bond