Peptides sequence

Balance data

Type	Literature	Year	Data	Acc		URL
Anticancer peptides	Anticancer peptides prediction with deep representation learning features	2021	Download ↓ Main Dataset: P-861/ N-861 Alternate Dataset: P-970/ N-970	5-CV Main Dataset: 79.1% Alternate Dataset: 94.5%	Independent testing Main Dataset: 77.5% Alternate Dataset: 93.0%	iACP-DRLF
	AntiCP 2.0: an updated model for predicting anticancer peptides	2021		5-CV Main Dataset: 75.29% Alternate Dataset: 90.10%	Independent testing Main Dataset: 75.43% Alternate Dataset: 92.01%	AntiCP 2.0
	ACPred-Fuse: fusing multi-view information improves the prediction of anticancer peptides	2020	Download ↓ Training dataset: ACP-125; non-ACPs-125; Independent test set: ACPs-82; non-ACPs-2628;	10-CV 82.4%	Independent testing 89.0%	ACPred-Fuse
	ACPred-FL: a sequence-based predictor using effective feature representation to improve the prediction of anti-cancer peptides	2018	Download ↓ Training dataset (ACP500): P-250/ N-250 Testing dataset (ACP164): P-970/ N-970	5-CV ACP500: 91.4%	Independent testing ACP164: 88.4%	ACPred-FL
Anti-hypertensive peptides	Boosted Prediction of Antihypertensive Peptides Using Deep Learning	2021	Download ↓ Training dataset: AHTPs: 913; non-AHTPs: 913; Independent test sets: AHTPs: 386; non-AHTPs: 386;	10-CV 95%	Independent testing 88.9%	/
Anti-hypertensive peptides	mAHTPred: a sequence-based meta-predictor for improving the prediction of anti-hypertensive peptides using effective feature representation	2019		10-CV 0.848	Independent testing 0.883	mAHTPred
Anti-Tubercular Peptides	AtbPpred: A Robust Sequence-Based Prediction of Anti-Tubercular Peptides Using Extremely Randomized Trees	2019	Download ↓ AntiTb_MD Training dataset: AtbPs: 199; non-AtbPs: 199; Independent test sets: AtbPs: 47; non-AtbPs: 47; AntiTb_RD Training dataset: AtbPs: 199; non-AtbPs: 199; Independent test sets: AtbPs: 47; non-AtbPs: 47;	10-CV AntiTb_MD 0.849 AntiTb_RD 0.917	Independent testing AntiTb_MD 0.894 AntiTb_RD 0.851	AtbPpred
Therapeutic peptides	ITP-Pred: an interpretable method for predicting, therapeutic peptides with fused features low-dimension representation	2021	Download ↓ QSP400: Training dataset: P-400/ N-400; Testing dataset: P-40/ N-40 CPP740: Training dataset: P-740/ N-740; Testing dataset: P-92/ N-92	5-CV QSP400: 87.0% CPP740: 87.3%	Independent testing QSP400: 95.1% CPP740: 97.5%	/
	PPTPP: a novel therapeutic peptide prediction method using physicochemical property encoding and adaptive feature representation learning	2020	Download ↓ Training dataset: AAP: P-107/ N-107; ABP: P-800/ N-800; ACP: P-250/ N-250; AIP: P-1258/ N-1887; AVP: P-544/ N-407; CPP: P-370/ N-370; QSP: P-200/ N-200; SBP: P-80/ N-80; Independent test sets: Main: AAP: P-28/ N-28; ABP: P-199/ N-199; ACP: P-82/ N-82; AIP: P-420/ N-629; AVP: P-60/ N-54; CPP: P-92/ N-92; QSP: P-20/ N-20; SBP: P-24/ N-20; Alternative: AAP: P-28/ N-2000; ABP: P-199/ N-2000; ACP: P-82/ N-2000; AIP: P-420/ N-2000; AVP: P-60/ N-2000; CPP: P-92/ N-2000; QSP: P-20/ N-2000; SBP: P-24/ N-2000;	10-CV AAP: AUC 0.871; ABP: AUC 0.977; ACP: AUC 0.927; AIP: AUC 0.772; AVP: AUC 0.930; CPP: AUC 0.975; QSP: AUC 0.972; SBP: AUC 0.865;	Independent testing Main AUC: AAP: 0.770; ABP: 0.988; ACP: 0.883; AIP: 0.720; AVP: 0.946; CPP: 0.965; QSP: 0.944; SBP: 0.740;	PPTPP
	PEPred-Suite: improved and robust prediction of therapeutic peptides using adaptive feature representation learning	2019		10-CV AAP: AUC 0.874; ABP: AUC 0.976; ACP: AUC 0.950; AIP: AUC 0.778; AVP: AUC 0.924; CPP: AUC 0.972; QSP: AUC 0.968; SBP: AUC 0.813;	Independent testing Main AUC: AAP: 0.804; ABP: 0.976; ACP: 0.949; AIP: 0.751; AVP: 0.949; CPP: 0.952; QSP: 0.960; SBP: 0.679; Alternative AUC: AAP: 0.774; ABP: 0.969; ACP: 0.638; AIP: 0.638; AVP: 0.945; CPP: 0.878; QSP: 0.897; SBP: 0.796;	PEPred-Suite
Toxic peptides	ATSE: a peptide toxicity predictor by exploiting structural and evolutionary information based on graph neural network and attention mechanism	2021	Download ↓ Dataset: toxic peptides: 1932; non-toxic peptides:1932; (training set (85%) and test data (15%))	10-CV 0.952	Independent testing NA	ATSE
Cell-penetrating Peptides	TargetCPP: accurate prediction of cell‑penetrating peptides from optimized multi‑scale features using gradient boost decision tree	2020	Download ↓ CPP924: CPPs: 462; Non-CPPs: 462; CPPind: CPPs: 111; Non-CPPs: 34;	jackknife test CPP924: 93.45%	Independent testing CPPind: 88.28%	/
	StackCPPred: a stacking and pairwise energy content-based prediction of cell-penetrating peptides and their uptake efficiency	2020	Download ↓ CPP924: CPPs: 462; Non-CPPs: 462 CPPsite3: CPPs: 187; Non-CPPs: 187	jackknife test CPP924: 94.5%; CPPsite3: 78.3%		StackCPPred
	CPPred-FL: a sequence-based predictor for large-scale identification of cell-penetrating peptides by feature representation learning	2018		10-CV CPP924: 92.1%		CPPred-FL
	SkipCPP-Pred: an improved and promising sequence-based predictor for predicting cell-penetrating peptides	2018		jackknife test CPP924: 90.6%		SkipCPP-Pred
	CPPred-RF: A Sequence-based Predictor for Identifying Cell-Penetrating Peptides and Their Uptake Efficiency	2017		jackknife test CPP924: 91.6%; CPPsite3: 71.1%		CPPred-RF
Hemolytic peptide	HLPpred-Fuse: improved and robust prediction of hemolytic peptide and its activity by fusing multiple feature representation	2020	Download ↓ Dataset_1: Training dataset: P-433/ N-666; Independent test sets: P-423/ N-1999; Dataset_2: Training dataset: P-433/ N-106; Independent test sets: P-352/ N-92; Dataset_3: Training dataset: P-671/ N-168; Independent test sets: P-559/ N-147;	10-CV Dataset_1: 98.4%; Dataset_2: 81.3%; Dataset_3: 82.9%;	Independent testing Dataset_1: BACC: 0.905; Dataset_2: Acc:0.808; Dataset_3: BACC: 0.792;	HLPpred-Fuse
	A Method for Predicting Hemolytic Potency of Chemically Modified Peptides From Its Structure	2020	Download ↓ Training dataset: hemolytic peptides: 466; non hemolytic peptides: 466; Independent test sets: hemolytic peptides: 117; non hemolytic peptides: 117;	5-CV 78.3%	Independent testing 78.29%	HemoPlmod
	HAPPENN is a novel tool for hemolytic activity prediction for therapeutic peptides which employs neural networks	2020	Download ↓ Hemolytic peptides: 1543 Non-hemolytic peptides: 2195	10-CV 85.66%±1.93		HAPPENN
Bitter Peptide	iBitter-Fuse: A Novel Sequence-Based Bitter Peptide Predictor by Fusing Multi-View Features	2021	Download ↓ Training dataset: bitter peptides: 256; non-bitter peptides: 256; Independent test sets: bitter peptides: 64; non-bitter peptides: 64;	10-CV 0.918	Independent testing 0.930	iBitter-Fuse
	BERT4Bitter: a bidirectional encoder representations from transformers (BERT)-based model for improving the prediction of bitter peptides	2021		10-CV 0.816	Independent testing 0.922	BERT4Bitter
	iBitter-SCM: Identi fication and characterization of bitter peptides using a scoring card method with propensity scores of dipeptides	2020		10-CV 87.11%	Independent testing 84.38%	iBitter-SCM

Note: 5-CV：5-fold cross validation; 10-CV: 10-fold cross validation; P: Positive samples; N: Negative samples;
^∗Web-server/Code is not available at the time of writing

Classification based on Peptide sequence