D-I-TASSER P62714-D1

D-I-TASSER results for P62714-D1

[Click result.zip to download all results on this page]

Input Sequence in FASTA format

>P62714 (309 residues)
MDDKAFTKELDQWVEQLNECKQLNENQVRTLCEKAKEILTKESNVQEVRCPVTVCGDVHG
QFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYPERITILRGNHES
RQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHI
RALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRA
HQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHV
TRRTPDYFL

Predicted Secondary Structure

Sequence	20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| MDDKAFTKELDQWVEQLNECKQLNENQVRTLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYPERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFL
Prediction	CCCCHHHHHHHHHHHHHHHCCCCCHHHHHHHHHHHHHHHHHCCCSSSSCCCSSSSCCCCCCHHHHHHHHHHCCCCCCCCSSSCCCCCCCCCCCHHHHHHHHHHHHHCCCSSSSCCCCCCCCHHHCCCCHHHHHHHHHCCHHHHHHHHHHHHHCCCSSSSCCCSSSSCCCCCCCCCCHHHHHHCCCCCCCCCCCCCSSSSSCCCCCCCCCCCCCCCCSSSSCHHHHHHHHHHCCCCSSSSSCSCCCCCSSSSCCCSSSSSSCCCCCCCCCCCCSSSSSSCCCCCSSSSSSSCCCCCCCCCCCCCCCCCCC
Confidence	997313799999999998189999999999999999999869976984588599616888899999999866999988567237656776640799999999999789818861357643012122342388999728889999999999747402573790898469877788778999857899889877843356743898889963268997587299999999999099779870210212115632991899972687677889727999987998478999717899888888889997779
	H:Helix; S:Strand; C:Coil

Predicted Solvent Accessibility

Sequence	20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| MDDKAFTKELDQWVEQLNECKQLNENQVRTLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYPERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFL
Prediction	866642363045005303745404361024005202500472211130414020000000002000100332240453210124121321320000000000000211410110232441242233310223024212132003000100000000010343021234323331411420440613341243202000000003546414314000001002400230065150410020230053023122543100000243134424010000102771414032040245675453655326127
	Values range from 0 (buried residue) to 8 (highly exposed residue)

Predicted Contact, Hydrogen and Distance Map Used in D-I-TASSER simulation

Contact Map

Distance Map

Hydrogen bond networks

D-I-TASSER simulation is guided by the consensus contact map (left figure), distance map (middle figure) and Hydrogen bond network (right figure) derived based on confidence scores of DeepPotential. In the contact, distance map and hydrogen bond networks, the axes mark the residue index along the sequence. For the contact map, each dot represents a residue pair with predicted contact, while for the distance map and hydrogen bond network, a color scale represents a distance of 1-20+ angstroms or a angle of 0-180 degree.

Top 10 threading templates used by D-I-TASSER

Rank

PDB
hit

ID1

ID2

Cov

Norm.
Zscore

Download
alignment

                  20                  40                  60                  80                 100                 120                 140                 160                 180                 200                 220                 240                 260                 280                 300

                   |                   |                   |                   |                   |                   |                   |                   |                   |                   |                   |                   |                   |                   |                   |

SS
Seq

CCCCHHHHHHHHHHHHHHHCCCCCHHHHHHHHHHHHHHHHHCCCSSSSCCCSSSSCCCCCCHHHHHHHHHHCCCCCCCCSSSCCCCCCCCCCCHHHHHHHHHHHHHCCCSSSSCCCCCCCCHHHCCCCHHHHHHHHHCCHHHHHHHHHHHHHCCCSSSSCCCSSSSCCCCCCCCCCHHHHHHCCCCCCCCCCCCCSSSSSCCCCCCCCCCCCCCCCSSSSCHHHHHHHHHHCCCCSSSSSCSCCCCCSSSSCCCSSSSSSCCCCCCCCCCCCSSSSSSCCCCCSSSSSSSCCCCCCCCCCCCCCCCCCC
MDDKAFTKELDQWVEQLNECKQLNENQVRTLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYPERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFL

2nppC

0.97

0.94

26.21

1.50

DEthreader

-DEKVFTKELDQWIEQLNECKQLSESQVKSLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYRERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEP-----------

2nppC

0.97

27.21

3.54

SPARKS-K

-DEKVFTKELDQWIEQLNECKQLSESQVKSLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYRERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFL

2nppC

0.98

0.92

25.76

1.55

MapAlign

---KVFTKELDQWIEQLNECKQLSESQVKSLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYRERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPR---------------

2nppC

0.97

27.21

1.07

CEthreader

-DEKVFTKELDQWIEQLNECKQLSESQVKSLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYRERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFL

2nppC

0.97

27.21

3.28

MUSTER

-DEKVFTKELDQWIEQLNECKQLSESQVKSLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYRERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFL

5jjtA

0.37

10.95

3.86

HHsearch

IEGEVTLDFVKTMMEDFKNQKTLHKRYAYQIVLQTRQILLALPSLVDIGKHITVCGDVHGQFYDLLNIFELNGLPSEENYLFNGDFVDRGSFSVEIILTLFAFKCMCPSSIYLARGNHESKSMNKIYGFEGEVRSKLS-EKFVDLFAEVFCYLPLAHVINGKVFVVHGGLSVDGVKLSDIRAIDRFCEPPEEGLMCELLWSDPQPLPGRGPSKRGVGLSFGGDVTKRFLQDNNLDLLVRSHEVKDEGYEVEHDGKLITVFSAPNYCDQMGNKGAFIRFEADMKPNIVTFSAVPHPDVK-PMAYALRMF-

2nppC

0.97

27.21

3.48

FFAS-3D

-DEKVFTKELDQWIEQLNECKQLSESQVKSLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYRERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFL

2nppC

0.97

27.21

1.88

EigenThreader

-DEKVFTKELDQWIEQLNECKQLSESQVKSLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVDRGYYSVETVTLLVALKVRYRERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFL

2iaeC

0.97

0.96

26.94

3.15

CNFpred

---KVFTKELDQWIEQLNECKQLSESQVKSLCEKAKEILTKESNVQEVRCPVTVCGDVHGQFHDLMELFRIGGKSPDTNYLFMGDYVNRGYYSVETVTLLVALKVRYRERITILRGNHESRQITQVYGFYDECLRKYGNANVWKYFTDLFDYLPLTALVDGQIFCLHGGLSPSIDTLDHIRALDRLQEVPHEGPMCDLLWSDPDDRGGWGISPRGAGYTFGQDISETFNHANGLTLVSRAHQLVMEGYNWCHDRNVVTIFSAPNYCYRCGNQAAIMELDDTLKYSFLQFDPAPRRGEPHVTRRTPDYFM

1auiA

0.41

0.40

11.71

1.50

DEthreader

AFDNDGKPRVDILKAHLMKEGRLEESVALRIITEGASILRQEKNLLDIDAPVTVCGDIHGQFFDLMKLFEVGGSPANTRYLFLGDYVDRGYFSIECVLYLWALKILYPKTLFLLRGNHECRHLTEYFTFKQECKIKY-SERVYDACMDAFDCLPLAALMNQQFLCVHGGLSPEINTLDDIRKLDRFKEPPAYGPMCDILWSDPLEDFHFTHNTRGCSYFYSYPAVCEFLQHNNLLSILRAHEAQDAGYRMYRKPSLITIFSAPNYLDVYNNKAAVLKYENN-VMNIRQFNCS-PHPY--------SFEE

(a)	ID1 is the number of template residues identical to query divided by number of aligned residues.
(b)	ID2 is the number of template residues identical to query divided by query sequence length.
(c)	Cov is equal the number of aligned template residues divided by query sequence length.
(d)	Norm. Zscore is the normalized Z-score of the threading alignments. A Normalized Z-score >1 means a good alignment and is highlighted in bold.
(e)	Download alignment lists the threading program used to identify the template, and provide the 3D structure of aligned regions of threading templates (threading[1-10].pdb.gz).
(f)	Template residues identical to query sequence are highlighted in color.

Top 1 final models from D-I-TASSER

Click to view	Rank^a	Download	Estimated TM-score^b
	1	model1.pdb.gz	0.9034

(a)

D-I-TASSER simulations generate a large ensemble of structural conformations, i.e. decoys. These decoys are clustered by SPICKER based on pairwise structure similarity to report up to five final models from the five largest clusters. Models are ranked in descending order of cluster size. If the simulations converge well, it is possible to have less than 5 models generated, which is usually an indication of good model quality.

(b)

The model confidence is quantitatified by estimated TM-score (eTM-score), calculated based on significance of threading template alignments, contact map satisfaction rate, mean absolute error between distance of model and distance of DeepPotential, and convergence of D-I-TASSER simulations. eTM-score is typically in the range of [0, 1], with higher eTM-score signifies higher model confidence.

Proteins with similar structure

Top 10 structural analogs in PDB (as identified by TM-align)

Rank	PDB Hit	TM-score	RMSD^a	IDEN^a	Cov.	Download Alignment
1	2nppC	0.99	0.66	0.974	0.997	model1_2nppC.pdb.gz
2	6tz6A	0.92	2.05	0.365	0.974	model1_6tz6A.pdb.gz
3	1auiA	0.92	2.21	0.407	0.977	model1_1auiA.pdb.gz
4	6tz8A	0.92	2.03	0.381	0.971	model1_6tz8A.pdb.gz
5	1s70A	0.92	1.49	0.450	0.945	model1_1s70A.pdb.gz
6	5jpeA	0.92	1.56	0.423	0.945	model1_5jpeA.pdb.gz
7	1wao12	0.91	1.93	0.373	0.958	model1_1wao12.pdb.gz
8	5jjtA	0.91	1.94	0.365	0.958	model1_5jjtA.pdb.gz
9	3icfA	0.87	1.85	0.348	0.906	model1_3icfA.pdb.gz
10	2z72A2	0.67	2.96	0.133	0.764	model1_2z72A2.pdb.gz

(a)	Query structure is shown in cartoon, while the structural analog is displayed using backbone trace.
(b)	Ranking of proteins is based on TM-score of the structural alignment between the query structure and known structures in the PDB library.
(c)	RMSD^a is the RMSD between residues that are structurally aligned by TM-align.
(d)	IDEN^a is the percentage sequence identity in the structurally aligned region.
(e)	Cov. represents the coverage of the alignment by TM-align and is equal to the number of structurally aligned residues divided by length of the query protein.

Predicted Gene Ontology (GO) Terms

Molecular Function (MF)

GO term	Cscore^GO	Name
GO:0003824	0.73	catalytic activity
GO:0016787	0.69	hydrolase activity
GO:0016791	0.68	phosphatase activity
GO:0004721	0.66	phosphoprotein phosphatase activity
GO:0004722	0.59	protein serine/threonine phosphatase activity

Download full result of the above consensus prediction.

Click the graph to show a high resolution version.

(a)

Cscore^GO is the confidence score of predicted GO terms. Cscore^GO values range in between [0-1]; where a higher value indicates a better confidence in predicting the function using the template.

(b)

The graph shows the predicted terms within the Gene Ontology hierachy for Molecular Function. Confidently predicted terms are color coded by Cscore^GO:

[0.4,0.5)

[0.5,0.6)

[0.6,0.7)

[0.7,0.8)

[0.8,0.9)

[0.9,1.0]

Biological Process (BP)

GO term	Cscore^GO	Name
GO:0009987	0.90	cellular process
GO:0065007	0.70	biological regulation
GO:0050789	0.68	regulation of biological process
GO:0008152	0.67	metabolic process
GO:0044237	0.66	cellular metabolic process
GO:0071704	0.65	organic substance metabolic process
GO:0044238	0.65	primary metabolic process
GO:0043170	0.64	macromolecule metabolic process
GO:0044260	0.63	cellular macromolecule metabolic process
GO:0050794	0.57	regulation of cellular process
GO:0044699	0.56	single-organism process
GO:0019538	0.56	protein metabolic process
GO:0006796	0.55	phosphate-containing compound metabolic process
GO:0006464	0.54	cellular protein modification process
GO:0016311	0.52	dephosphorylation
GO:0006470	0.50	protein dephosphorylation

Download full result of the above consensus prediction.

Click the graph to show a high resolution version.

(a)

Cscore^GO is the confidence score of predicted GO terms. Cscore^GO values range in between [0-1]; where a higher value indicates a better confidence in predicting the function using the template.

(b)

The graph shows the predicted terms within the Gene Ontology hierachy for Biological Process. Confidently predicted terms are color coded by Cscore^GO:

[0.4,0.5)

[0.5,0.6)

[0.6,0.7)

[0.7,0.8)

[0.8,0.9)

[0.9,1.0]

Cellular Component (CC)

GO term	Cscore^GO	Name
GO:0044464	1.00	cell part
GO:0044424	0.97	intracellular part
GO:0043226	0.70	organelle
GO:0043229	0.67	intracellular organelle
GO:0032991	0.67	macromolecular complex
GO:0043234	0.65	protein complex
GO:0043227	0.62	membrane-bounded organelle
GO:1902494	0.59	catalytic complex
GO:0043231	0.59	intracellular membrane-bounded organelle
GO:0044444	0.58	cytoplasmic part
GO:0008287	0.58	protein serine/threonine phosphatase complex
GO:0005634	0.50	nucleus

Download full result of the above consensus prediction.

Click the graph to show a high resolution version.

(a)

Cscore^GO is the confidence score of predicted GO terms. Cscore^GO values range in between [0-1]; where a higher value indicates a better confidence in predicting the function using the template.

(b)

The graph shows the predicted terms within the Gene Ontology hierachy for Cellular Component. Confidently predicted terms are color coded by Cscore^GO:

[0.4,0.5)

[0.5,0.6)

[0.6,0.7)

[0.7,0.8)

[0.8,0.9)

[0.9,1.0]

Predicted Enzyme Commission (EC) Numbers

Top 5 enzyme homologs in PDB

Rank	Cscore^EC	PDB Hit	TM-score	RMSD^a	IDEN^a	Cov.	EC Number	Predicted Active Site Residues
1	0.513	1auiA	0.921	2.21	0.407	0.977	3.1.3.16	88,89,117,118,214,241
2	0.488	1g5bB	0.561	2.95	0.165	0.644	3.1.3.16	88,118
3	0.362	2z72A	0.695	3.48	0.132	0.819	3.1.3.48	88,89,117,118,241
4	0.338	1s95A	0.911	1.84	0.371	0.955	3.1.3.16	125,195
5	0.219	2nppC	0.987	0.66	0.974	0.997	3.1.3.16	88,89,117,118,214,241

	Click on the radio buttons to visualize predicted active site residues.
(a)	Cscore^EC is the confidence score for the Enzyme Commission (EC) number prediction. Cscore^EC values range in between [0-1]; where a higher score indicates a more reliable EC number prediction.
(b)	TM-score is a measure of global structural similarity between query and template protein.
(c)	RMSD^a is the RMSD between residues that are structurally aligned by TM-align.
(d)	IDEN^a is the percentage sequence identity in the structurally aligned region.
(e)	Cov. represents the coverage of global structural alignment and is equal to the number of structurally aligned residues divided by length of the query protein.

Predicted Ligand Binding Sites

Template proteins with similar binding site:

Rank	Cscore^LB	PDB Hit	TM-score	RMSD^a	IDEN^a	Cov.	BS-score	Lig. Name	Download Complex	Predicted binding site residues
1	0.86	2nppC	0.987	0.66	0.974	0.997	1.92	III	complex1.pdb.gz	89,123,127,189,190,191,200,214,215,243,265,266,269
2	0.85	2nylF	0.939	0.67	0.973	0.948	1.57	MN	complex2.pdb.gz	57,59,85,118
3	0.77	3hvqA	0.903	1.12	0.495	0.922	1.55	PO4	complex3.pdb.gz	59,85,89,117,118,214,241
4	0.58	1it6B	0.903	0.92	0.489	0.919	1.54	CYU	complex4.pdb.gz	89,122,123,127,189,190,191,200,203,204,212,213,214,216,219,242,243,265,269
5	0.50	1it6B	0.903	0.92	0.489	0.919	1.84	MN	complex5.pdb.gz	85,117,167,241
6	0.50	2ie41	0.925	0.56	0.979	0.932	1.92	III	complex6.pdb.gz	51,67,70,71,73,74,76,77,78,79,106,107,110,277,280,287,293
7	0.48	2o8g0	0.904	1.16	0.486	0.926	1.53	III	complex7.pdb.gz	42,43,45,47,48,49,79,89,109,110,112,118,122,123,125,126,127,160,162,191,214,216,235,236,241,243,245,248,250,254,265,266,268,269,280,281,282,283,284,285,286,288
8	0.45	3h63A	0.913	2.01	0.377	0.961	1.56	NHC	complex8.pdb.gz	57,59,85,89,117,118,167,214,241,242,243,260,265
9	0.36	1u32A	0.902	0.96	0.482	0.919	1.62	OKA	complex9.pdb.gz	89,118,123,127,213,214,216,265,266,269
10	0.08	3ll8A	0.922	2.17	0.407	0.977	1.30	III	complex10.pdb.gz	67,71,248,249,254,272,278,282,283,284,285,286,287,288,290

	Click on the radio buttons to visualize predicted binding site and residues.
(a)	Cscore^LB is the confidence score of predicted binding site. Cscore^LB values range in between [0-1]; where a higher score indicates a more reliable ligand-binding site prediction.
(b)	BS-score is a measure of local similarity (sequence & structure) between template binding site and predicted binding site in the query structure. Based on large scale benchmarking analysis, we have observed that a BS-score >1 reflects a significant local match between the predicted and template binding site.
(c)	TM-score is a measure of global structural similarity between query and template protein.
(d)	RMSD^a the RMSD between residues that are structurally aligned by TM-align.
(e)	IDEN^a is the percentage sequence identity in the structurally aligned region.
(f)	Cov. represents the coverage of global structural alignment and is equal to the number of structurally aligned residues divided by length of the query protein.

[Click result.zip to download all results on this page]

References:
1.	Wei Zheng, Qiqige Wuyun, Yang Li, Quancheng Liu, Xiaogen Zhou, Yiheng Zhu, P. Lydia Freddolino, Yang Zhang. Integrating deep learning potentials with I-TASSER for single- and multi-domain protein structure prediction. Submitted. 2023.