Composition

Protein-level¶

In PyPropel, the function fpseq governs the extraction of features from the whole proteins. At the protein level, PyPropel considers compositions of amino acids across the whole protein sequence. It can be done below.

AAC¶

AAC stands for the amino acid composition.

Python

import pypropel as pp

aac = pp.fpseq.composition(
    seq=seq,
    mode='aac',
)

Output

{'A': 0.119403, 'C': 0.074627, 'D': 0.149254, 'E': 0.089552, 'F': 0.059701, 'G': 0.089552, 'H': 0.0, 'I': 0.0, 'K': 0.014925, 'L': 0.029851, 'M': 0.044776, 'N': 0.029851, 'P': 0.029851, 'Q': 0.029851, 'R': 0.014925, 'S': 0.119403, 'T': 0.014925, 'V': 0.044776, 'W': 0.029851, 'Y': 0.014925}

DAC¶

DAC stands for the di-amino acid composition.

Python

import pypropel as pp

dac = pp.msa.composition(
    seq=seq,
    mode='dac',
)

Output

[['AA', 0.015152], ['AC', 0.0], ['AD', 0.075758], ..., ['YW', 0.0], ['YY', 0.0]]

TAC¶

TAC stands for the tri-amino acid composition.

Python

import pypropel as pp

tac = pp.msa.composition(
    seq=seq,
    mode='tac',
)

Output

[['AAA', 0.0], ['AAC', 0.0], ['AAD', 0.015385], ...]

QAC¶

QAC stands for the qua-amino acid composition.

Python

import pypropel as pp

qac = pp.msa.composition(
    seq=seq,
    mode='qac',
)

Output

[['AAAA', 0.0], ['AAAC', 0.0], ['AAAD', 0.0], ...]

CKSNAP¶

CKSNAP measures the average frequencies of every two amino acids that are k spaced.

Python

import pypropel as pp

cksnap = pp.msa.composition(
    seq=seq,
    k_spaced=3,
    mode='cksnap',
)

Output

{'AA': 0.047619, 'AC': 0, 'AD': 0, 'AE': 0.015873, 'AF': 0, 'AG': 0.015873, 'AH': 0, 'AI': 0, 'AK': 0, 'AL': 0, 'AM': 0, 'AN': 0, 'AP': 0, 'AQ': 0, 'AR': 0, 'AS': 0.031746, 'AT': 0, 'AV': 0, 'AW': 0, 'AY': 0, 'CA': 0, 'CC': 0.015873, 'CD': 0.015873, 'CE': 0.015873, 'CF': 0, 'CG': 0, 'CH': 0, 'CI': 0, 'CK': 0.015873, 'CL': 0, 'CM': 0, 'CN': 0, 'CP': 0, 'CQ': 0, 'CR': 0, 'CS': 0, 'CT': 0, 'CV': 0.015873, 'CW': 0, 'CY': 0, 'DA': 0.015873, 'DC': 0, 'DD': 0.031746, 'DE': 0.015873, 'DF': 0, 'DG': 0, 'DH': 0, 'DI': 0, 'DK': 0, 'DL': 0.015873, 'DM': 0, 'DN': 0, 'DP': 0, 'DQ': 0, 'DR': 0, 'DS': 0.031746, 'DT': 0, 'DV': 0.015873, 'DW': 0.015873, 'DY': 0, 'EA': 0, 'EC': 0, 'ED': 0.015873, 'EE': 0, 'EF': 0.031746, 'EG': 0, 'EH': 0, 'EI': 0, 'EK': 0, 'EL': 0.015873, 'EM': 0, 'EN': 0, 'EP': 0, 'EQ': 0.015873, 'ER': 0, 'ES': 0, 'ET': 0, 'EV': 0, 'EW': 0, 'EY': 0, 'FA': 0.015873, 'FC': 0, 'FD': 0.031746, 'FE': 0, 'FF': 0, 'FG': 0, 'FH': 0, 'FI': 0, 'FK': 0, 'FL': 0, 'FM': 0, 'FN': 0.015873, 'FP': 0, 'FQ': 0, 'FR': 0, 'FS': 0, 'FT': 0, 'FV': 0, 'FW': 0, 'FY': 0, 'GA': 0, 'GC': 0, 'GD': 0, 'GE': 0, 'GF': 0, 'GG': 0.063492, 'GH': 0, 'GI': 0, 'GK': 0, 'GL': 0, 'GM': 0.015873, 'GN': 0.015873, 'GP': 0, 'GQ': 0, 'GR': 0, 'GS': 0, 'GT': 0, 'GV': 0, 'GW': 0, 'GY': 0, 'HA': 0, 'HC': 0, 'HD': 0, 'HE': 0, 'HF': 0, 'HG': 0, 'HH': 0, 'HI': 0, 'HK': 0, 'HL': 0, 'HM': 0, 'HN': 0, 'HP': 0, 'HQ': 0, 'HR': 0, 'HS': 0, 'HT': 0, 'HV': 0, 'HW': 0, 'HY': 0, 'IA': 0, 'IC': 0, 'ID': 0, 'IE': 0, 'IF': 0, 'IG': 0, 'IH': 0, 'II': 0, 'IK': 0, 'IL': 0, 'IM': 0, 'IN': 0, 'IP': 0, 'IQ': 0, 'IR': 0, 'IS': 0, 'IT': 0, 'IV': 0, 'IW': 0, 'IY': 0, 'KA': 0.015873, 'KC': 0, 'KD': 0, 'KE': 0, 'KF': 0, 'KG': 0, 'KH': 0, 'KI': 0, 'KK': 0, 'KL': 0, 'KM': 0, 'KN': 0, 'KP': 0, 'KQ': 0, 'KR': 0, 'KS': 0, 'KT': 0, 'KV': 0, 'KW': 0, 'KY': 0, 'LA': 0, 'LC': 0, 'LD': 0, 'LE': 0, 'LF': 0, 'LG': 0, 'LH': 0, 'LI': 0, 'LK': 0, 'LL': 0, 'LM': 0, 'LN': 0, 'LP': 0, 'LQ': 0, 'LR': 0, 'LS': 0.031746, 'LT': 0, 'LV': 0, 'LW': 0, 'LY': 0, 'MA': 0, 'MC': 0, 'MD': 0, 'ME': 0, 'MF': 0, 'MG': 0, 'MH': 0, 'MI': 0, 'MK': 0, 'ML': 0, 'MM': 0.015873, 'MN': 0, 'MP': 0, 'MQ': 0, 'MR': 0, 'MS': 0, 'MT': 0, 'MV': 0, 'MW': 0.015873, 'MY': 0.015873, 'NA': 0, 'NC': 0, 'ND': 0, 'NE': 0, 'NF': 0.015873, 'NG': 0, 'NH': 0, 'NI': 0, 'NK': 0, 'NL': 0, 'NM': 0.015873, 'NN': 0, 'NP': 0, 'NQ': 0, 'NR': 0, 'NS': 0, 'NT': 0, 'NV': 0, 'NW': 0, 'NY': 0, 'PA': 0, 'PC': 0.015873, 'PD': 0, 'PE': 0, 'PF': 0, 'PG': 0, 'PH': 0, 'PI': 0, 'PK': 0, 'PL': 0, 'PM': 0, 'PN': 0, 'PP': 0, 'PQ': 0.015873, 'PR': 0, 'PS': 0, 'PT': 0, 'PV': 0, 'PW': 0, 'PY': 0, 'QA': 0, 'QC': 0.031746, 'QD': 0, 'QE': 0, 'QF': 0, 'QG': 0, 'QH': 0, 'QI': 0, 'QK': 0, 'QL': 0, 'QM': 0, 'QN': 0, 'QP': 0, 'QQ': 0, 'QR': 0, 'QS': 0, 'QT': 0, 'QV': 0, 'QW': 0, 'QY': 0, 'RA': 0, 'RC': 0, 'RD': 0, 'RE': 0, 'RF': 0, 'RG': 0, 'RH': 0, 'RI': 0, 'RK': 0, 'RL': 0, 'RM': 0, 'RN': 0, 'RP': 0, 'RQ': 0, 'RR': 0, 'RS': 0.015873, 'RT': 0, 'RV': 0, 'RW': 0, 'RY': 0, 'SA': 0.015873, 'SC': 0, 'SD': 0.015873, 'SE': 0.031746, 'SF': 0, 'SG': 0, 'SH': 0, 'SI': 0, 'SK': 0, 'SL': 0, 'SM': 0, 'SN': 0, 'SP': 0.015873, 'SQ': 0, 'SR': 0, 'SS': 0.015873, 'ST': 0, 'SV': 0.015873, 'SW': 0, 'SY': 0, 'TA': 0, 'TC': 0, 'TD': 0, 'TE': 0, 'TF': 0, 'TG': 0, 'TH': 0, 'TI': 0, 'TK': 0, 'TL': 0, 'TM': 0, 'TN': 0, 'TP': 0.015873, 'TQ': 0, 'TR': 0, 'TS': 0, 'TT': 0, 'TV': 0, 'TW': 0, 'TY': 0, 'VA': 0, 'VC': 0, 'VD': 0, 'VE': 0.015873, 'VF': 0.015873, 'VG': 0, 'VH': 0, 'VI': 0, 'VK': 0, 'VL': 0, 'VM': 0, 'VN': 0, 'VP': 0, 'VQ': 0, 'VR': 0, 'VS': 0, 'VT': 0.015873, 'VV': 0, 'VW': 0, 'VY': 0, 'WA': 0, 'WC': 0, 'WD': 0.015873, 'WE': 0, 'WF': 0, 'WG': 0, 'WH': 0, 'WI': 0, 'WK': 0, 'WL': 0, 'WM': 0, 'WN': 0, 'WP': 0, 'WQ': 0, 'WR': 0.015873, 'WS': 0, 'WT': 0, 'WV': 0, 'WW': 0, 'WY': 0, 'YA': 0, 'YC': 0, 'YD': 0.015873, 'YE': 0, 'YF': 0, 'YG': 0, 'YH': 0, 'YI': 0, 'YK': 0, 'YL': 0, 'YM': 0, 'YN': 0, 'YP': 0, 'YQ': 0, 'YR': 0, 'YS': 0, 'YT': 0, 'YV': 0, 'YW': 0, 'YY': 0}

AVEANF¶

The AVEANF feature reflects the information about positions and frequencies of amino acids in a protein sequence. For details, please refer to ¹

Python

import pypropel as pp

aveanf = pp.msa.composition(
    seq=seq,
    mode='aveanf',
)

Output

{'A': 0.22597633674152806, 'C': 0.1652824858757062, 'D': 0.17214478688616622, 'E': 0.09744508222164794, 'F': 0.0459034244741305, 'G': 0.4204055204055204, 'H': 0, 'I': 0, 'K': 0.045454545454545456, 'L': 0.03737373737373738, 'M': 0.10492898913951544, 'N': 0.04793028322440087, 'P': 0.055322128851540614, 'Q': 0.059848484848484845, 'R': 0.017857142857142856, 'S': 0.08590396237795747, 'T': 0.1, 'V': 0.10087719298245613, 'W': 0.04096989966555184, 'Y': 0.04}

MSA-level¶

First, we can read a MSA of protein 1aijL as follows.

Python

import pypropel as pp

msa = pp.msa.read(
    msa_fpn=to('data/msa/aln/1aijL.aln'),
)

AAC¶

Then, the AAC of MSA is calculated below.

Python

import pypropel as pp

aac = pp.msa.composition(
    msa=msa,
    mode='aac',
)

Output

{'A': 0.05826085292688736, 'C': 0.004700385606685579, 'D': 0.008600878716107662, 'E': 0.015672130729309737, 'F': 0.05294113895278523, 'G': 0.06534156201388167, 'H': 0.019759785927038455, 'I': 0.04494023459261235, 'K': 0.0027040633087310126, 'L': 0.05968535215833517, 'M': 0.022540825417799226, 'N': 0.01639108826946766, 'P': 0.03155605592517563, 'Q': 0.011760641022454832, 'R': 0.017077715813173088, 'S': 0.03935616241622517, 'T': 0.02184298134471232, 'V': 0.040838283818220966, 'W': 0.01871466930702302, 'Y': 0.0222107075628659, '-': 0.42443809036607894}

Site-level¶

There is no method to calculate compositions of amino acids because no accurate einformation can be reflected from a site itself in a protein sequence given only.

Zhen Chen, Pei Zhao, Fuyi Li, Tatiana T Marquez-Lago, André Leier, Jerico Revote, Yan Zhu, David R Powell, Tatsuya Akutsu, Geoffrey I Webb, Kuo-Chen Chou, A Ian Smith, Roger J Daly, Jian Li, Jiangning Song, iLearn: an integrated platform and meta-learner for feature engineering, machine-learning analysis and modeling of DNA, RNA and protein sequence data, Briefings in Bioinformatics, Volume 21, Issue 3, May 2020, Pages 1047–1057, https://doi.org/10.1093/bib/bbz041 ↩