silenus092/AmpsConverter

AMPsConverter

(Version 04.04.2021)

Tools for fasta conversion that supports the following AMP Databases

AVPdb database of experimentally validated antiviral peptides
HIPdb A Database of Experimentally Validated HIV Inhibiting Peptide
AntiBP2 : Server for antibacterial peptide prediction
DBAASP V.3 Database of Antimicrobial Activity and Structure of Peptides
DADP: Database of Anuran Defense Peptides
CancerPPD Database of Anticancer peptides and proteins
AntiTbPred | Prediction of antitubercular peptides
AntiCP Consists of 225 antimicrobial peptides with anticancer properties.
Antifp | is an in silico method, which is developed to predict and design antifungal peptides
ADAM | A Database of Anti-Microbial peptides
Inverpred | is a specialized database of AMPs from invertebrates.
PeptideDB database assembles all naturally occurring signalling peptides from animal source
BACTIBASE : Database Dedicated to Bacteriocin (Filter by antimicrobial keyword )
PeptideDB database assembles all naturally occurring signalling peptides from animal source
Antimicrobial Peptide Database (APD3, https://wangapd3.com/downloads.php)
MilkAMP A database for milk AMPs (a comprehensive database of antimicrobial peptides of dairy origin)
BACTIBASE : Database Dedicated to Bacteriocin (Filter by antimicrobial keyword )
AVPpred: collection and prediction of highly effective antiviral peptides.
ADAPTABLE is both a webserver and data-miner of antimicrobial peptides. 
BaAMPs | Biofilm-active AMPs database: antimicrobial peptides (AMPs) specifically tested against microbial biofilms.
dbAMP v1.4, an integrated resource for exploring antimicrobial peptides
DRAMP 2.0, an updated data repository of antimicrobial peptides
Hemolytik Database of Hemolytic activity of AMPs. <-- เอา ID ของ  CP and toxix ออก
EROP-Moscow The EROP-Moscow oligopeptide database
EnzyBase Database of enzybiotics (lysins, autolysins, lysozymes, and large bacteriocins) (อันที่เป็น ช่องว่างในต่ำเเหน่งที่สองของ ID ถือว่าเป็น synthetic)
AmPEP: Sequence-based prediction of antimicrobial 
-----
Not include:
YADAMP: yet another database of antimicrobial peptides  

Collecting and Fitlering

Strategies:

1. Convert lower case to upper case
2. Remove white space between seqeunce or beginning and ending (e.g. "VEWNEMT WMEWEREI ENYTKLIYKILEESQEQ" to "VEWNEMTWMEWEREIENYTKLIYKILEESQEQ")
3. Remove empty lines
4. Remove none protien seqeunce charater or unrelated words (e.g. "+FRPKFGK-" or "structure given" or "NOT AVAILABLE")
5. Remove unsure sequence. (e.g. "fCYwO-CyLeu-Pen-TKKr" or "βAlaFTicO")

Databases (Versions: Sep 01.2020)

Global variable and function

root_path = "/mnt/c/works/RKI/AMP_DB/AMPs"
root_output_path = "/mnt/c/works/RKI/AMP_DB/fasta"

def createID(ID, source, seq_type=None, description=None):
    header = str(">" + ID + "|" + source)
    if seq_type is not None:
        header = header + "|" + seq_type
    if description is not None:
        header = header + "|" + description
    return header 

DBAASP

How To download original source

1. Visit https://dbaasp.org/search
2. Click "Download CSV" at the bottom of page

How To convert originial source to FASTA format

print("--- dbaasp_peptides ----")
filename = "dbaasp_peptides"
input_path = os.path.join(root_path, filename+".csv")
out_path = os.path.join(root_output_path, filename+".fasta")

df = pd.read_csv(input_path, header=0, usecols=['ID', 'SEQUENCE'])
# create new ID
df['ID'] = df['ID'].apply(lambda row: str(row)+"_dbaasp")

# Group duplication
df['ID'] = df['ID'].astype(str)
grouped_df = df.groupby("SEQUENCE")
grouped_lists = grouped_df["ID"].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        header = createID(row['ID'], filename)
        # print(header)
        file.write(header + '\n')
        seq = row['SEQUENCE']
        # print(seq)
        file.write(seq + '\n')

print("--- End of dbaasp_peptides ----")

Note for original resource:

1. lower case

AVPdb

How To download original source

1. Visit http://crdd.osdd.net/servers/avpdb/download_data.php
2. Select Natural peptides

How To convert originial source to FASTA format

print("--- AVPdb_data ----")
filename = "AVPdb_data"
input_path = os.path.join(root_path, filename+".tsv")
out_path = os.path.join(root_output_path, filename+".fasta")
# Load File
df = pd.read_csv(input_path, header=[0], delimiter='\t')

# Group duplication
df['Id'] = df['Id'].astype(str)
grouped_df = df.groupby("Sequence")
grouped_lists = grouped_df["Id"].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(file_name)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        header = createID(row['Id'], filename)
        # print(header)
        file.write(header + '\n')
        seq = row['Sequence']
        # print(seq)
        file.write(seq + '\n')
print("--- End of AVPdb_data ----")

HIPdb

How To download original source

1. Visit http://crdd.osdd.net/servers/hipdb/download_data.php

How To convert originial source to FASTA format

print("--- HIPdb_data ----")
filename = "HIPdb_data"
input_path = os.path.join(root_path, filename+".csv")
out_path = os.path.join(root_output_path, filename+".fasta")

df = pd.read_csv(input_path, header=0, encoding="utf-8", usecols=['ID', 'SEQUENCE'])

# Group duplication
df['ID'] = df['ID'].astype(str)
grouped_df = df.groupby("SEQUENCE")
grouped_lists = grouped_df["ID"].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        header = createID(row['ID'], filename)
        # print(header)
        file.write(header + '\n')
        seq = row['SEQUENCE']
        # print(seq)
        file.write(seq + '\n')
print("--- End of HIPdb_data ----")

Note for original resource:

1. file input format (utf-8, utf-16)

Hemolytik

How To download original source

1. Download from http://crdd.osdd.net/raghava/hemolytik/allsequences.txt

How To convert originial source to FASTA format

print("--- Hemolytik_allsequences ----")
filename = "Hemolytik_allsequences_02_09_2020"
input_path = os.path.join(root_path, filename+".txt")
out_path = os.path.join(root_output_path, filename+".fasta")

df = pd.read_csv(input_path, header=[0], sep='\t', encoding='utf8' )
# create new ID
df['ID'] = df['ID'].apply(lambda row: str(row)+"_Hemolytik_allsequences")

searchfor = ["β", "[", "]","(" , ")", "/", "-", "Ψ","Δ",
 "1","2","3","4","5","6","7","8","9","0"] 

# Group duplication
df['ID'] = df['ID'].astype(str)
grouped_df = df.groupby("SEQ")
grouped_lists = grouped_df["ID"].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        seq = row['SEQ']
        if any(x in seq for x in searchfor):
            print("special:"+seq)
            continue
        if any(c.islower() for c in seq):
            seq = seq.upper()
            # print("Upper:"+seq)

        if " " in seq:
            print("Found whitespace:"+seq)
            seq = seq.replace(" ", "")
        header = createID(row['ID'], filename)
        # print(header)
        file.write(header + '\n')
        
        # print(seq)
        file.write(seq + '\n')

print("--- End of Hemolytik_allsequences ----")

Note:

1. missing char? X-NH-GVATQLLAAYILLFDEYNEKKASAQKDILIKVL
2. lower case?
3. βAlaFTicOicβAlaKTicOicβAlaFTicOicβAlaKTicKKKK ?

BaAmps

How To download original source

1. Access to this website http://www.baamps.it/browse and select ALL in dropdown, then click export.

How To convert originial source to FASTA format

print("--- BAAMPs_data ----")
filename = "BAAMPs_data"
input_path = os.path.join(root_path, filename+".csv")
out_path = os.path.join(root_output_path, filename+".fasta")
# Load File
df = pd.read_csv(input_path, header=[0])
# remove duplicate among same ID
df = df.drop_duplicates(subset=["PeptideID", "PeptideName", "PeptideSequence"], keep="last")
# Group duplication
df['PeptideID'] = df['PeptideID'].astype(str)
grouped_df = df.groupby("PeptideSequence")
grouped_lists = grouped_df["PeptideID"].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(file_name)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        header = createID(row['PeptideID'], filename)
        # print(header)
        file.write(header + '\n')
        seq = row['PeptideSequence']
        # print(seq)
        file.write(seq + '\n')
print("--- End of BAAMPs_data ----")

Note:

1. lower case?

MilkAMP

How To

1. Download from http://milkampdb.org/entrieslist.php?export=csv

How To convert originial source to FASTA format

print("--- milkampdb ----")
filename = "milkampdb"
input_path = os.path.join(root_path, filename+".csv")
out_path = os.path.join(root_output_path, filename+".fasta")

df = pd.read_csv(input_path, header=0, usecols=['id', 'Sequence'])
# trim white space 
df['Sequence'] = df['Sequence'].str.strip()
# remove empty value / Not protein symbol
df = df.dropna()
df = df[~df.Sequence.str.contains("Not determined")]
# chop semicolon 
df['Sequence'] = df['Sequence'].str.split(';').str[0]

# Group duplication
grouped_df = df.groupby('Sequence')
grouped_lists = grouped_df['id'].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        header = createID(row['id'], filename)
        # print(header)
        file.write(header + '\n')
        seq = row['Sequence']
        # print(seq)
        file.write(seq + '\n')

print("--- End of milkampdb ----")

Note:

1. different format? APRKNVRWCTISQPEW/CIRA;

DADP

DADP: Database of Anuran Defense Peptides

How To

Access to the below link and save it as a json file

http://split4.pmfst.hr/dadp/?a=listAjax&classId=&search=undefined&limit=undefined

How To convert originial source to FASTA format

print("--- DAPD ----")
filename = "DADP"
input_path = os.path.join(root_path, filename+".json")
out_path = os.path.join(root_output_path, filename+".fasta")

df = pd.read_json(input_path)
# split column in to multiple columns by comma 
df = pd.concat([df['aaData'], df['aaData'].apply(pd.Series)],axis=1)
# Group duplication
df[0] = df[0].astype(str)
grouped_df = df.groupby(6)
grouped_lists = grouped_df[0].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()


print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        if row[6] == "/":
            continue
        header = createID(row[0], filename)
        # print(header)
        file.write(header + '\n')
        seq = row[6]
        # print(seq)
        file.write(seq + '\n')

print("--- End of DAPD ----")

CancerPPD

Database of Anticancer peptides and proteins

How To download original source

curl http://crdd.osdd.net/raghava/cancerppd/natural_seq_dwn/l_natural.txt -o CancerPPD_NaturalAminoAcids_l_natural.txt

curl http://crdd.osdd.net/raghava/cancerppd/natural_seq_dwn/d_natural.txt -o CancerPPD_NaturalAminoAcids_d_natural.txt

curl http://crdd.osdd.net/raghava/cancerppd/non-natural_seq_dwn/l_non-natural.txt -o CancerPPD_l_non-natural.txt

curl http://crdd.osdd.net/raghava/cancerppd/non-natural_seq_dwn/d_non-natural.txt -o CancerPPD_d_non-natural.txt

curl http://crdd.osdd.net/raghava/cancerppd/natural_seq_dwn/mix_natural.txt  -o CancerPPD_mix_natural.txt

curl http://crdd.osdd.net/raghava/cancerppd/non-natural_seq_dwn/mix_non-natural.txt  -o CancerPPD_mix_non-natural.txt

How To convert originial source to FASTA format

def converterForCancerPPD(filename):
    input_path = os.path.join(root_path, filename+".txt")
    out_path = os.path.join(root_output_path, filename+".fasta")
    # Load File
    df = pd.read_csv(input_path, sep='\t', header=[0])
    df = groupCancerPPDDuplication(df)
    with open(out_path, 'w') as file:
        for index, row in df.iterrows():
            header = createID(str(row['id']), filename)
            #print(header)
            file.write(header + '\n')
            seq = row['Sequence'].upper()
            #print(seq)
            file.write(seq + '\n')

def groupCancerPPDDuplication(df):
    df['id'] = df['id'].astype(str)
    grouped_df = df.groupby("Sequence")
    grouped_lists = grouped_df["id"].agg(lambda column: ",".join(column))
    grouped_lists = grouped_lists.reset_index(name="id")
    return grouped_lists

file_names=["CancerPPD_d_natural",
            "CancerPPD_l_natural",
            "CancerPPD_mix_natural",
            ]

print("--- CancerPPD ----")

for file_name in file_names:
    print(file_name)
    converterForCancerPPD(file_name)

print("--- End of CancerPPD ----")

AntiCP

Consists of 225 antimicrobial peptides with anticancer properties.

How To download original source

1. Access to the below URL and then looks for "Main Dataset"

https://webs.iiitd.edu.in/raghava/anticp/datasets.php

2. Then click "P" under Donwload column which will redirect to https://webs.iiitd.edu.in/raghava/anticp/datasets/225_amp_pos.txt

Note:

1. They also provide alternate datasets (e.g., potential non-anticancer peptides , without anticancer activities)

How To convert originial source to FASTA format

print("--- anticp_225_amp_pos ----")
filename = "anticp_225_amp_pos"
input_path = os.path.join(root_path, filename+".txt")
out_path = os.path.join(root_output_path, filename+".fasta")

# create new ID
df = pd.read_csv(input_path,  names=["Sequence"], sep='\t')
df['Id'] = df.index
df['Id'] = df['Id'].apply(lambda row: str(row)+"_anticp_225")

# Group duplication
df['Id'] = df['Id'].astype(str)
grouped_df = df.groupby("Sequence")
grouped_lists = grouped_df["Id"].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        header = createID(row['Id'], filename)
        # print(header)
        file.write(header + '\n')
        seq = row['Sequence']
        # print(seq)
        file.write(seq + '\n')
print("--- End of anticp_225_amp_pos ----")

Antimicrobial Peptide Database (APD3)

How To download original source

# http://aps.unmc.edu/AP/downloads.php
wget http://aps.unmc.edu/AP/APD3_update2020_release.fasta

How To convert originial source to FASTA format

print("--- APD3_update2020_release ----")
filename = "APD3_update2020_release"
input_path = os.path.join(root_path, filename+".fasta")
out_path = os.path.join(root_output_path, filename+".fasta")

with open(input_path, mode='r') as in_file, \
     open(out_path, mode='w') as out_file:
    for line in in_file:
        if line.startswith(">"): # skip lines that start with > 
            out_file.write(line)
            continue
        removedWhite_line = line.strip() # strip whitespace
        out_file.write(removedWhite_line + '\n')

print("--- End of APD3_update2020_release ----")

EROP-Moscow

How To download original source

1. Go to thise page http://erop.inbi.ras.ru/query-erop.php?
2. Under functional class, select antimicrobial

or Download from the link below

http://erop.inbi.ras.ru/result1.php?EROP_NMB_K=&PEP_NAME_K=&FAM_NAME_K=&ALL_KAR__K=&ALL_KGD__K=&ALL_PHYL_K=&ALL_B_CL_K=&Organism=&SPECIES__K=&ALL_TISS_K=&SEQ_1____K=&AAR_SUM__K1=&AAR_SUM__K2=&M_W______K1=&M_W______K2=&PI_______K1=&PI_______K2=&POSIT_CH_K1=&POSIT_CH_K2=&NEGAT_CH_K1=&NEGAT_CH_K2=&NET_CH___K1=&NET_CH___K2=&FUNC_CL__K=antimicrobial&FUNCTION_K=&SEQ_REFA_V=&SEQ_REFT_V=&SEQ_REFJ_V=&YEAR_SEQ_V1=&YEAR_SEQ_V2=&COUNTRY__V=&page_mode=Download_Fasta 

How To convert originial source to FASTA format

# erop
# +: to denote +H2, which is the open N-terminus
# b: for an acetyl residue or other chemical group at the N-terminus
# -: to denote O-, which is the open C-terminus
# z: for an amide bond at the C-terminus
# J: to denote the pyroglutaminyl linkage, formed by an N-terminal glutamine, due to side-chain reaction with the terminal amine residue
# U: for the (occasional) aminoisobutyric acid residue.
#
print("--- erop ----")
filename = "erop"
input_path = os.path.join(root_path, filename+".FASTA")
out_path = os.path.join(root_output_path, filename+".fasta")

with open(input_path, mode='r') as in_file, \
     open(out_path, mode='w') as out_file:
    for line in in_file:
        if line.startswith(">"): # skip lines that start with > 
            out_file.write(line.strip()+ '\n')
            continue

        line = line.replace("+", "")
        line = line.replace("-", "")
        line = line.replace("b", "")
        line = line.replace("z", "")
        if " " in line:
            line = line.replace(" ", "")
        out_file.write(line)

print("--- End of erop ----")

Note:

1. special character?

LAMP2

How To download original source

1. Download from http://biotechlab.fudan.edu.cn/database/lamp/db/lamp2.fasta

How To convert originial source to FASTA format

print("--- LAMP2 ----")
filename = "LAMP2"
input_path = os.path.join(root_path, filename+".FASTA")
out_path = os.path.join(root_output_path, filename+".fasta")

with open(input_path, mode='r') as in_file, \
     open(out_path, mode='w') as out_file:
    for line in in_file:
        if line.startswith(">"): # skip lines that start with > 
            out_file.write(line.strip()+ '\n')
            continue
        if " " in line:
            line = line.replace(" ", "")
        line = line.upper()
        out_file.write(line)
    out_file.write("\n")

print("--- End of LAMP2 ----")

Note:

1. lower case?

BACTIBASE

How To download original source

1. Access to http://bactibase.hammamilab.org/bacteriocinslist.php, and then input the condition on the search bar (e.g. antimicrobial)
2. Select show "all" at the bottom of the page.
3. Click "FASTA" icon to download

How To convert originial source to FASTA format

1. It already offers FASTA format

EnzyBase

How To download original source

1. Downlaod from http://biotechlab.fudan.edu.cn/database/enzybase/db/enzy2.fasta

How To convert originial source to FASTA format

print("--- enzy2 ----")
from Bio import *

filename = "enzy2"
input_path = os.path.join(root_path, filename+".fasta")
out_path = os.path.join(root_output_path, filename+".fasta")

with open(out_path, 'w') as f_out:
    for seq_record in SeqIO.parse(open(input_path, mode='r'), 'fasta'):
        if "-" in  seq_record.seq:
            print('SequenceID = '  + seq_record.id)
            continue
        r=SeqIO.write(seq_record, f_out, 'fasta')
        if r!=1: print('Error while writing sequence:  ' + seq_record.id)

print("--- End of enzy2 ----")

DRAMP (2.0)

How To

1. Vist the http://dramp.cpu-bioinfor.org/downloads/
2. Click download link

How To convert originial source to FASTA format

ef converterForDRAMP(filename):
    input_path = os.path.join(root_path, filename+".xlsx")
    out_path = os.path.join(root_output_path, filename+".fasta")
    # Load File
    df = pd.read_excel(input_path, header=[0])
    print("size:"+ str(len(df.index)))
    df = groupDRAMPDuplication(df)
    with open(out_path, 'w') as file:
        for index, row in df.iterrows():
            if "unknown" in  row['Sequence'].lower():
                continue
            if  "-spacer-" in  row['Sequence']:
                continue
            header = createID(row['DRAMP_ID'], filename)
            #print(header)
            file.write(header + '\n')
            seq = row['Sequence'] 
            if " " in seq:
                #print("Found whitespace:"+seq)
                seq = seq.replace(" ", "")
            if "-" in seq:
                print("Found '-':"+seq)
            file.write(seq + '\n')

def groupDRAMPDuplication(df):
    df['DRAMP_ID'] = df['DRAMP_ID'].astype(str)
    grouped_df = df.groupby("Sequence")
    grouped_lists = grouped_df["DRAMP_ID"].agg(lambda column: ",".join(column))
    grouped_lists = grouped_lists.reset_index()
    return grouped_lists

file_names=["DRAMP_Antibacterial_amps",
            "DRAMP_Anticancer_amps",
            "DRAMP_Antifungal_amps",
            "DRAMP_Anti-Gram-_amps",
            "DRAMP_Anti-Gram+_amps",
            "DRAMP_Antimicrobial_amps",
            "DRAMP_Antiparasitic_amps",
            "DRAMP_Antiprotozoal_amps",
            "DRAMP_Antitumor_amps",
            "DRAMP_Antiviral_amps",
            "DRAMP_Insecticidal_amps",
            ]

print("--- DRAMP ----")

for file_name in file_names:
    print(file_name)
    converterForDRAMP(file_name)

print("--- End of DRAMP ----")

AVPpred

How To

1. Vist http://crdd.osdd.net/servers/avppred/collection.php
2. Click download link
3. Fix header (e.g. "AVP ID" change to "AVP_ID")
4. some sequences may splited into serveral lines (5-6 sequences), manully fix them.

How To convert originial source to FASTA format

print("--- AVPPred ----")

filename = "AVPPred"
input_path = os.path.join(root_path, filename+".tsv")
out_path = os.path.join(root_output_path, filename+".fasta")

df = pd.read_csv(input_path, header=0, usecols=['AVP_ID', 'Sequence'], comment='#', sep="\t")
# trim white space 
df['Sequence'] = df['Sequence'].str.strip()
# remove empty value / Not protein symbol
df = df.dropna()

# Group duplication
grouped_df = df.groupby('Sequence')
grouped_lists = grouped_df['AVP_ID'].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        header = createID(row['AVP_ID'], filename)
        # print(header)
        file.write(header + '\n')
        # print(seq)
        file.write(row['Sequence'] + '\n')
print("--- End of AVPPred ----")

PeptideDB

How To

1. Visit http://www.peptides.be/index.php?p=search&accession_number=&name=&organism_group=&organism_species=&length_from=&length_to=&mass_from=&mass_to=&family_group=Antimicrobial+peptides&family=&uniprot=&aminoacid=
2. Then click "Peptides in FASTA"

How To convert originial source to FASTA format

print("--- peptideDB.anti ----")

filename = "peptideDB.anti"

counter=0
input_path = os.path.join(root_path, filename+".FASTA")
out_path = os.path.join(root_output_path, filename+".fasta")
df = pd.DataFrame(columns=['ID','Sequence'])

with open(input_path, mode='r') as in_file:
    for line in in_file:
        if line in ['\n', '\r\n', ""]:
            continue
        ID = str(counter)+"_"+filename
        counter+=1
        new_row = {'ID':ID, 'Sequence':line}
        df = df.append(new_row, ignore_index=True)

# Group duplication
grouped_df = df.groupby('Sequence')
grouped_lists = grouped_df['ID'].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        seq = row['Sequence']

        header = createID(row['ID'], filename)
        # print(header)
        file.write(header + '\n')
        # print(seq)
        file.write(seq )

print("--- End of peptideDB.anti ----")

dbAMPv1.4

How To

1. Visit http://140.138.77.240/~dbamp/download.php
2. Click download under "Download All Antimicrobial peptides Sequence Data" panel

How To convert originial source to FASTA format

print("--- dbAMPv1.4_validated ----")
filename = "dbAMPv1.4_validated"
input_path = os.path.join(root_path, filename+".FASTA")
out_path = os.path.join(root_output_path, filename+".fasta")

with open(input_path, mode='r') as in_file, \
     open(out_path, mode='w') as out_file:
    for line in in_file:
        if line.startswith(">"): # skip lines that start with > 
            out_file.write(line.strip()+ '\n')
            continue
        line = line.upper()
        out_file.write(line)

print("--- End of dbAMPv1.4_validated ----")

Inverpred

How To

1. Get the original source via Email.
2. Manully extract records and save into .txt format

How To convert originial source to FASTA format

print("--- inverpred  ----")
filename = "inverpred"
input_path = os.path.join(root_path, filename+".txt")
out_path = os.path.join(root_output_path, filename+".fasta")
df = pd.read_csv(input_path, header=0, sep="\t")
# create new ID
df['ID'] = df.index
df['ID'] = df['ID'].apply(lambda row: str(row)+"_"+filename)

# Group duplication
grouped_df = df.groupby('Secuencia')
grouped_lists = grouped_df['ID'].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()
print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        seq = row['Secuencia']
        header = createID(row['ID'], filename)
        # print(header)
        file.write(header + '\n')
        # print(seq)
        file.write(seq + '\n' )
print("--- End of inverpredd ----")

Antifp

How To

1. Visit https://webs.iiitd.edu.in/raghava/antifp/algo.php
2. Donwload data from Antifp_Main selection (select only positive set)
3. Merge from two sources manaully and save into .txt format

How To convert originial source to FASTA format

print("--- Antifp pos_train_ds3  ----")

filename = "pos_train_ds3"
input_path = os.path.join(root_path, filename+".txt")
out_path = os.path.join(root_output_path, filename+".fasta")
df = pd.read_csv(input_path, header=0, sep="\t")
# create new ID
df['ID'] = df.index
df['ID'] = df['ID'].apply(lambda row: str(row)+"_"+filename)

# Group duplication
grouped_df = df.groupby('Sequence')
grouped_lists = grouped_df['ID'].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()
print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        seq = row['Sequence']
        header = createID(row['ID'], filename)
        # print(header)
        file.write(header + '\n')
        # print(seq)
        file.write(seq + '\n' )
print("--- End of Antifp pos_train_ds3 ----")

ADAM

How To

1. Visit http://bioinformatics.cs.ntou.edu.tw/adam/search_amp.html
2. Click "show all" button
3. Manually save record from the result table into .txt and then fixs newline problem (scan through 7000 records, troublesome!!!)

How To convert originial source to FASTA format

print("---  ADAM  ----")

filename = "adam_test"
input_path = os.path.join(root_path, filename+".txt")
out_path = os.path.join(root_output_path, filename+".fasta")
df = pd.read_csv(input_path, header=0, sep="\t")

# Group duplication
grouped_df = df.groupby('Sequence')
grouped_lists = grouped_df['ID'].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()

print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        seq = row['Sequence']
        header = createID(row['ID'], filename)
        # print(header)
        file.write(header + '\n')
        # print(seq)
        file.write(seq + '\n' )
print("--- End of ADAM ----")

AntiTbPred

How To

1. Visit https://webs.iiitd.edu.in/raghava/antitbpred/down.php
2. Download only first row (Positive dataset)

How To convert originial source to FASTA format

print("--- antitbpred  ----")

filename = "antitbpred"
input_path = os.path.join(root_path, filename+".txt")
out_path = os.path.join(root_output_path, filename+".fasta")
df = pd.read_csv(input_path, header=0, sep="\t")
# create new ID
df['ID'] = df.index
df['ID'] = df['ID'].apply(lambda row: str(row)+"_"+filename)

# Group duplication
grouped_df = df.groupby('Sequence')
grouped_lists = grouped_df['ID'].agg(lambda column: ",".join(column))
grouped_lists = grouped_lists.reset_index()
print(filename)
with open(out_path, 'w') as file:
    for index, row in grouped_lists.iterrows():
        seq = row['Sequence']
        header = createID(row['ID'], filename)
        # print(header)
        file.write(header + '\n')
        # print(seq)
        file.write(seq + '\n' )
print("--- End of antitbpred ----")

Clustering (CD-HIT)

We use CD-HIT for clustering sequences to reduce sequence redundancy

./cd-hit -i /mnt/c/works/RKI/AMPsConverter/AMP_DB/one_fasta_file.sorted.fasta -o /mnt/c/works/RKI/AMPsConverter/AMP_DB/stats/nr100 -c 1  

grep ">" /mnt/c/works/RKI/AMPsConverter/AMP_DB/stats/nr100 | wc -l 

Statistics

Report basic statistics , please see statistics.py

1. Report amino acid character in file
2. Report lenght distribution
3. Save a graph report in PNG format.

---

Training set2

Under trainingset2 folder contains all raw sequences that are used for the machine learning task.