Examples 75%

This is a manual analysis of some of the Alphafold2 models that got a WHAT IF LR-score of ~75.
Again, the LR-score tells nothing about the protein analysed itself. The score only warns when a protein structure would be a useless template for a homology modelling experiment.
Very correct natively unfolded structures, very small structures, structures that consist mainly of one very long helix, etc, all get a high LR-score despite normally being the best representation of the protein at hand.

AF-A0A1B0GVY2-F1-model_v1.pdb

COMPND   2 MOLECULE: UNCHARACTERIZED PROTEIN;

MDFSLGLRLGPRNKKPAHQEPPACSRHGPPAGSPCLSCPPSACAFPCHGCPPPPCSCTAC
PSSAAPCPACPSLPGPPCTCSCPRCPACPPLTCPHNSCVSCSGPPLTCCHPSPCPVYPCS
MSRAACLSSRLGCSDSCSCGQGAAWGPPGSLGCCSCCFRGQQRTSRGRCLII

BLAST finds cysteine-rich proteins and keratin associated proteins that have many short repeats. The 25% that is structured in this model is found in a small domain and a single hairpin both in the bottom left. I also see one lone Cys-Cys bond somewhere. Many cysteines are in close proximity but not bridged and surface located.

AF-O95359-F1-model_v1.pdb

COMPND   2 MOLECULE: TRANSFORMING ACIDIC COILED-COIL-CONTAINING PROTEIN 2;

MGNENSTSDNQRTLSAQTPRSAQPPGNSQNIKRKQQDTPGSPDHRDASSIGSVGLGGFCT
ASESSASLDPCLVSPEVTEPRKDPQGARGPEGSLLPSPPPSQEREHPSSSMPFAECPPEG
CLASPAAAPEDGPQTQSPRREPAPNAPGDIAAAFPAERDSSTPYQEIAAVPSAGRERQPK
EEGQKSSFSFSSGIDQSPGMSPVPLREPMKAPLCGEGDQPGGFESQEKEAAGGFPPAESR
QGVASVQVTPEAPAAAQQGTESSAVLEKSPLKPMAPIPQDPAPRASDRERGQGEAPPQYL
TDDLEFLRACHLPRSNSGAAPEAEVNAASQESCQQPVGAYLPHAELPWGLPSPALVPEAG
GSGKEALDTIDVQGHPQTGMRGTKPNQVVCVAAGGQPEGGLPVSPEPSLLTPTEEAHPAS
SLASFPAAQIPIAVEEPGSSSRESVSKAGMPVSADAAKEVVDAGLVGLERQVSDLGSKGE
HPEGDPGEVPAPSPQERGEHLNTEQSHEVQPGVPPPPLPKEQSHEVQPGAPPPPLPKAPS
ESARGPPGPTDGAKVHEDSTSPAVAKEGSRSPGDSPGGKEEAPEPPDGGDPGNLQGEDSQ
AFSSKRDPEVGKDELSKPSSDAESRDHPSSHSAQPPRKGGAGHTDGPHSQTAEADASGLP
HKLGEEDPVLPPVPDGAGEPTVPEGAIWEGSGLQPKCPDTLQSREGLGRMESFLTLESEK
SDFPPTPVAEVAPKAQEGESTLEIRKMGSCDGEGLLTSPDQPRGPACDASRQEFHAGVPH
PPQGENLAADLGLTALILDQDQQGIPSCPGEGWIRGAASEWPLLSSEKHLQPSQAQPETS
IFDVLKEQAQPPENGKETSPSHPGFKDQGADSSQIHVPVEPQEDNNLPTHGGQEQALGSE
LQSQLPKGTLSDTPTSSPTDMVWESSLTEESELSAPTRQKLPALGEKRPEGACGDGQSSR
VSPPAADVLKDFSLAGNFSRKETCCTGQGPNKSQQALADALEEGSQHEEACQRHPGASEA
ADGCSPLWGLSKREMASGNTGEAPPCQPDSVALLDAVPCLPALAPASPGVTPTQDAPETE
ACDETQEGRQQPVPAPQQKMECWATSDAESPKLLASFPSAGEQGGEAGAAETGGSAGAGD
PGKQQAPEKPGEATLSCGLLQTEHCLTSGEEASTSALRESCQAEHPMASCQDALLPAREL
GGIPRSTMDFSTHQAVPDPKELLLSGPPEVAAPDTPYLHVDSAAQRGAEDSGVKAVSSAD
PRAPGESPCPVGEPPLALENAASLKLFAGSLAPLLQPGAAGGEIPAVQASSGSPKARTTE
GPVDSMPCLDRMPLLAKGKQATGEEKAATAPGAGAKASGEGMAGDAAGETEGSMERMGEP
SQDPKQGTSGGVDTSSEQIA

Swissprot tells us: Transforming acidic coiled-coil-containing protein 2, and that it contains many, long disordered regions. The bottom right had side contains some structured bits, but these are merely extended loops crossing each other.

AF-Q96EU6-F1-model_v1.pdb

COMPND   2 MOLECULE: RIBOSOMAL RNA PROCESSING PROTEIN 36 HOMOLOG;

MPGANYRAGAGAGAGARRPRGARDREEDGGGLEPAAVARDLLRGTSNMSFEELLELQSQV
GTKTYKQLVAGNSPKKQASRPPIQNACVADKHRPLEMSAKIRVPFLRQVVPISKKVARDP
RFDDLSGEYNPEVFDKTYQFLNDIRAKEKELVKKQLKKHLSGEEHEKLQQLLQRMEQQEM
AQQERKQQQELHLALKQERRAQAQQGHRPYFLKKSEQRQLALAEKFKELKRSKKLENFLS
RKRRRNAGKDRRHLPLSKE

SwissProt:
function Involved in the early processing steps of the pre-rRNA in the maturation pathway leading to the 18S rRNA. {ECO:0000269|PubMed:20038530}.
subcellular location Nucleus, nucleolus {ECO:0000269|PubMed:20038530}. Note=Concentrated in the fibrillar region of the nucleolus.
It is very basic, so I can imagine it wraps around RNA backbones, or something similar that doesn′t require much structure. It also contains several long disordered stretches.
The model holds several helices, but where they pack they do it in strange ways.

AF-O95567-F1-model_v1.pdb

COMPND   2 MOLECULE: UNCHARACTERIZED PROTEIN C22ORF31;

MHPINVRRDPSIPIYGLRQSILLNTRLQDCYVDSPALTNIWMARTCAKQNINAPAPATTS
SWEVVRNPLIASSFSLVKLVLRRQLKNKCCPPPCKFGEGKLSKRLKHKDDSVMKATQQAR
KRNFISSKSKQPAGHRRPAGGIRESKESSKEKKLTVRQDLEDRYAEHVAATQALPQDSGT
AAWKGRVLLPETQKRQQLSEDTLTIHGLPTEGYQALYHAVVEPMLWNPSGTPKRYSLELG
KAIKQKLWEALCSQGAISEGAQRDRFPGRKQPGVHEEPVLKKWPKLKSKK

One third of this protein is disordered. I see two small domains (one of two packed helices, and one of 4 strands in a single sheet).

AF-Q8WV24-F1-model_v1.pdb

COMPND   2 MOLECULE: PLECKSTRIN HOMOLOGY-LIKE DOMAIN FAMILY A MEMBER 1;

MRRAPAAERLLELGFPPRCGRQEPPFPLGVTRGWGRWPIQKRREGARPVPFSERSQEDGR
GPAARSSGTLWRIRTRLSLCRDPEPPPPLCLLRVSLLCALRAGGRGSRWGEDGARLLLLP
PARAAGNGEAEPSGGPSYAGRMLESSGCKALKEGVLEKRSDGLLQLWKKKCCILTEEGLL
LIPPKQLQHQQQQQQQQQQQQQQQPGQGPAEPSQPSGPAVASLEPPVKLKELHFSNMKTV
DCVERKGKYMYFTVVMAEGKEIDFRCPQDQGWNAEITLQMVQYKNRQAILAVKSTRQKQQ
HLVQQQPPSQPQPQPQLQPQPQPQPQPQPQPQSQPQPQPQPKPQPQQLHPYPHPHPHPHS
HPHSHPHPHPHPHPHQIPHPHPQPHSQPHGHRLLRSTSNSA

Contains several disordered regions and several low complexity regions. The structured bit of this model holds a few strands from which two long helices protrude.