[ AlignSS | ToxScore ]

AlignSS : sequence alignment guided by secondary structure and accessibility

AlignSS ( seq_1 seq_2 r_stickiness )

pairwise sequence alignment that takes into account two additional properties of sequence the secondary structure and the residue accessibility and allows to vary general 'stickiness' (0. by default) to prevent unalignment of dissimilar parts. The residue exposure (aka residue solvent accessibility) values are assigned to a sequence if it is extracted from a 3D object with secondary structure and pre-calculated surface area. Example:

read pdb "1crn" 
assign sstructure # makes correct assignment according to hydrogen bonds
show surface area
make sequence
show 1crn_a
 >1crn_a crambin; chain: a; engineered: yes
 TTCCPSIVAR SNFNVCRLPG TPEAICATYT GCIIIPGATC PGDYAN
 >
 EEEE__HHHH HHHHHHHHH_ __HHHHHHHH _EEEE_____ ______
 >#
 4100336502 4405513775 2534500761 3024194673 386254  # compressed accessibility number. 0 - buried, 9 fully exposed

The normal sequence alignment (see align or Align ) will then be biased according to two weights ( the R_weights array) for the alpha-alpha and beta-beta match accordingly. The default values, 1.6 and 1.8, were optimized on a structural alignment benchmark. The i_window parameter determines smoothing of the edges of the secondary structures with the Smooth function (see the code of the macro).

Examples:

read pdb "2rh1" # beta 2 adrenergic
read pdb "1f88" # rhodopsin
delete a_*.!1   # leave only the main chains
show area surface a_1.
show area surface a_2.
assign sstructure a_*.
make sequence a_*.1
ali3d = AlignSS( 2rh1_a 1f88_a 0. )

See also:

• align 3D : structure derived alignment
• Align
• Smooth

ToxScore

ToxScore(X,l_addCovPenalty (yes),l_addProblematicPropertyName (no))

Structural alert SMARTS strings were collected from various source (Chemdiv, Enamine, ochem, etc). Due to the difference in severity, we calculated each SMARTS string's frequency of occurrence in approved drugs and use the frequency and labelled endpoint (e.g promiscuity vs acute toxicity) to assign a tox demerit score for each SMARTS string. Covalent groups (such as beta-lactam) can be either flagged as undesirable or tolerated by the l_addCovPenalty option. Option l_addProblematicPropertyName will add a second column to the table with the names of problematic properties.

For any test compound, the toxscore is the summation of the tox demerit score for that compound. About 20% of approved drugs have toxscore > 1. A limit of 1.5 may be useful for selection of compounds for virtual screening for a non-covalent binder.

Example:

add column t Chemical( {"COOCC","CCC#N"} )
add column t ToxScore (t.mol yes yes )  # simple static addition
  # now a fancy dynamic one:
add column t function="ToxScore(mol,yes,yes)" name="TS" append vector