diff --git a/docs/results/screen_HDAC2_full.csv b/docs/results/screen_HDAC2_full.csv new file mode 100644 index 0000000..3063b6c --- /dev/null +++ b/docs/results/screen_HDAC2_full.csv @@ -0,0 +1,300 @@ +rank,drug,P_binder,affinity,inclusion_reason +1,trichostatin-a,0.9995425343513489,-2.8145930767059326,related_mechanism +2,vorinostat,0.9994102716445923,-1.736063003540039,related_mechanism +3,panobinostat,0.9985321760177612,-2.5815885066986084,related_mechanism +4,belinostat,0.9969940185546875,-2.1257152557373047,related_mechanism +5,BG-1003,0.9969587326049805,-1.7902384996414185,general_sample +6,scriptaid,0.9953755140304565,-1.604792833328247,related_mechanism +7,mocetinostat,0.9910330772399902,-1.1003236770629883,related_mechanism +8,entinostat,0.9900848865509033,-0.66916424036026,related_mechanism +9,apicidin,0.9881076812744141,-2.1283297538757324,related_mechanism +10,BRD-K14666757,0.9683372974395752,-0.6549662351608276,general_sample +11,JW55,0.9362592697143555,-0.4662295877933502,general_sample +12,valproic-acid,0.8981399536132812,0.37430477142333984,related_mechanism +13,FIT,0.8313940763473511,0.08768215775489807,general_sample +14,hydroxyurea,0.7578531503677368,1.246307611465454,ground_truth +15,ibuprofen,0.7412509918212891,2.238396167755127,related_mechanism +16,sulforaphane,0.739348292350769,0.3284870982170105,related_mechanism +17,BRD-K10302728,0.7058488130569458,-0.9450978636741638,general_sample +18,BRD-K61285042,0.6548928022384644,0.49822238087654114,general_sample +19,ciprofibrate,0.633934736251831,1.9368321895599365,general_sample +20,KI-16425,0.6012023091316223,0.6640514731407166,general_sample +21,curcumin,0.5989675521850586,1.486403226852417,related_mechanism +22,BRD-K29660619,0.5905249118804932,0.12171633541584015,general_sample +23,resveratrol,0.5783771276473999,1.2722764015197754,related_mechanism +24,BRD-K27824357,0.564523458480835,0.1906198263168335,general_sample +25,tadalafil,0.5615900754928589,0.007123969495296478,related_mechanism +26,BRD-K46241566,0.5496933460235596,0.8938988447189331,general_sample +27,BRD-K99633092,0.5473401546478271,-0.44378313422203064,general_sample +28,BRD-K32389477,0.5432510375976562,-0.9505497813224792,general_sample +29,talnetant,0.5365580320358276,-0.4235907196998596,general_sample +30,montelukast,0.5170086622238159,0.7837215662002563,related_mechanism +31,tetracaine,0.514898419380188,0.486468642950058,general_sample +32,SA-1480001,0.5129876732826233,-0.14114731550216675,general_sample +33,lawsone,0.5126880407333374,1.3275636434555054,general_sample +34,quercetin,0.5112301111221313,0.35112306475639343,related_mechanism +35,BRD-K67328514,0.47897130250930786,1.6453043222427368,general_sample +36,glutamine,0.4580795168876648,2.2292144298553467,ground_truth +37,romidepsin,0.438159316778183,-0.6780526638031006,related_mechanism +38,BRD-K19362120,0.43379056453704834,-0.38182151317596436,general_sample +39,rifaximin,0.43118005990982056,-0.007878035306930542,general_sample +40,palmitoylethanolamide,0.4253748655319214,1.5332996845245361,general_sample +41,D-64131,0.4243965148925781,1.8345916271209717,general_sample +42,deferasirox,0.414570152759552,0.8550795912742615,related_mechanism +43,BRD-K39379309,0.39516061544418335,1.2569950819015503,general_sample +44,cetirizine,0.39126813411712646,1.497683048248291,negative_control +45,decitabine,0.3905729651451111,0.9013112783432007,related_mechanism +46,ingenol,0.38832157850265503,1.2636841535568237,general_sample +47,BRD-K02558072,0.38382846117019653,1.3527982234954834,general_sample +48,lestaurtinib,0.37580811977386475,0.5706255435943604,general_sample +49,BRD-K22546959,0.373577356338501,1.4228181838989258,general_sample +50,sulfasalazine,0.3713478446006775,0.9489234089851379,related_mechanism +51,brompheniramine,0.36528241634368896,0.7533001899719238,general_sample +52,SKI-II,0.3633918762207031,1.0488131046295166,general_sample +53,BRD-K03142004,0.3538649380207062,0.4557762145996094,general_sample +54,miltefosine,0.3524576723575592,1.3234755992889404,general_sample +55,lidocaine,0.3505249321460724,2.832113265991211,negative_control +56,omeprazole,0.3403707444667816,1.4137210845947266,negative_control +57,BW-B70C,0.34008413553237915,1.2555820941925049,general_sample +58,azacitidine,0.33303695917129517,0.8310854434967041,related_mechanism +59,KO-143,0.33238181471824646,-0.9562602043151855,general_sample +60,trioxsalen,0.33212581276893616,2.046276569366455,general_sample +61,dexchlorpheniramine,0.3283562660217285,1.02620530128479,general_sample +62,ciprofloxacin,0.32668250799179077,0.26640501618385315,negative_control +63,aspirin,0.3254982829093933,2.339493751525879,related_mechanism +64,BRD-K55966666,0.3211105465888977,-0.2134701907634735,general_sample +65,carisoprodol,0.3141220510005951,1.6034519672393799,general_sample +66,tetracycline,0.31255578994750977,1.2801803350448608,negative_control +67,atorvastatin,0.3122830092906952,0.5527983903884888,related_mechanism +68,BRD-K19105325,0.3045218586921692,-0.2390911877155304,general_sample +69,BRD-A27640568,0.3026970326900482,1.965691328048706,general_sample +70,phenolphthalein,0.3001948595046997,0.9219919443130493,general_sample +71,BRD-K71182391,0.29757264256477356,0.5220360159873962,general_sample +72,BRD-K38061943,0.2891416549682617,0.14858180284500122,general_sample +73,BRD-K93445052,0.28817978501319885,0.7619435787200928,general_sample +74,halofantrine,0.28806358575820923,0.008057385683059692,general_sample +75,BRD-K16912444,0.28658565878868103,-0.0714123547077179,general_sample +76,BRD-K39431601,0.2835044860839844,-0.3456823229789734,general_sample +77,LY-341495,0.27582740783691406,0.44341373443603516,general_sample +78,tyrphostin-A9,0.2754404842853546,1.5046072006225586,general_sample +79,BRD-K87449690,0.26641374826431274,2.289837121963501,general_sample +80,BRD-K87720600,0.2651839256286621,0.2896602153778076,general_sample +81,fexofenadine,0.26415905356407166,1.2466082572937012,negative_control +82,BRD-K33284065,0.2630389928817749,0.6950460076332092,general_sample +83,ITE,0.257363498210907,1.146105170249939,general_sample +84,amoxicillin,0.25568023324012756,1.8630526065826416,negative_control +85,BRD-K45889441,0.25448158383369446,0.6300027966499329,general_sample +86,BRD-K52524742,0.25103557109832764,1.058586835861206,general_sample +87,methotrexate,0.24424436688423157,1.9891445636749268,general_sample +88,BRD-K67470788,0.24184325337409973,1.50880765914917,general_sample +89,ethinyl-estradiol,0.24175500869750977,1.410076379776001,negative_control +90,lisinopril,0.2400914877653122,1.5881357192993164,general_sample +91,BRD-K93411257,0.23802407085895538,-0.2541484236717224,general_sample +92,BRD-K71353154,0.23799994587898254,0.5835225582122803,general_sample +93,cyclazosin,0.23158416152000427,0.03154781460762024,general_sample +94,BRD-K74271701,0.23076415061950684,0.7160882353782654,general_sample +95,BRD-K05647501,0.2291751652956009,-0.06926438212394714,general_sample +96,BRD-K78636275,0.22891822457313538,0.7736002206802368,general_sample +97,ditolylguanidine,0.2276442050933838,1.7872586250305176,general_sample +98,BRD-K12683703,0.22574575245380402,-0.5894231200218201,general_sample +99,BRD-K42197878,0.22489723563194275,0.9392799139022827,general_sample +100,ranitidine,0.2243157923221588,1.0414822101593018,negative_control +101,BRD-K00812396,0.2204073816537857,1.7750279903411865,general_sample +102,SA-1921783,0.21941909193992615,0.7298216223716736,general_sample +103,BRD-K57756744,0.21836763620376587,1.2867426872253418,general_sample +104,oprozomib,0.2171071171760559,1.1919209957122803,general_sample +105,astemizole,0.2156534045934677,1.7033143043518066,negative_control +106,BRD-K01977318,0.21503891050815582,1.926210641860962,general_sample +107,neostigmine,0.20658856630325317,1.2066024541854858,general_sample +108,BRD-K86997557,0.2063712477684021,1.7264163494110107,general_sample +109,BRD-K07276035,0.20610132813453674,0.22332103550434113,general_sample +110,BRD-K42733037,0.20607754588127136,0.4705517292022705,general_sample +111,BRD-K88092133,0.20329439640045166,1.5231980085372925,general_sample +112,BRD-K07464613,0.19792267680168152,0.9103862047195435,general_sample +113,BRD-K03463440,0.19763417541980743,0.04750436544418335,general_sample +114,U-54494A,0.19699649512767792,0.6771597266197205,general_sample +115,BRD-K95995661,0.19657522439956665,0.9159876704216003,general_sample +116,BRD-K35234011,0.19374209642410278,1.016001582145691,general_sample +117,BRD-A13914662,0.1926717460155487,0.49716946482658386,general_sample +118,BRD-K87426499,0.19177326560020447,-0.050703004002571106,general_sample +119,BRD-K80747759,0.19016335904598236,-0.5651810765266418,general_sample +120,BRD-K23154136,0.18987241387367249,2.289987802505493,general_sample +121,BRD-K52971552,0.18887174129486084,0.9786604046821594,general_sample +122,BRD-K15259924,0.1882244050502777,1.4186360836029053,general_sample +123,BRD-K60703528,0.18806274235248566,0.40015730261802673,general_sample +124,BRD-K27744061,0.1859944462776184,2.612982749938965,general_sample +125,BRD-K19416218,0.18407133221626282,1.7156975269317627,general_sample +126,loratadine,0.18328134715557098,2.494800090789795,negative_control +127,BRD-K27711805,0.18229132890701294,0.4912426173686981,general_sample +128,BRD-K67727634,0.17958897352218628,-0.022484712302684784,general_sample +129,BRD-K08995626,0.17850513756275177,3.004214286804199,general_sample +130,BRD-K76321890,0.17720356583595276,-0.8629432320594788,general_sample +131,PD-0325901,0.1763416975736618,1.1065287590026855,general_sample +132,BRD-K92028820,0.1763390302658081,0.864924430847168,general_sample +133,BRD-K58096890,0.17460352182388306,1.0638632774353027,general_sample +134,clotrimazole,0.17379066348075867,1.719211459159851,negative_control +135,terbinafine,0.1734161376953125,1.7575925588607788,negative_control +136,BRD-K82925070,0.17324930429458618,2.892310619354248,general_sample +137,BRD-K36820092,0.17034576833248138,0.4257924556732178,general_sample +138,BRD-K99628932,0.1687491536140442,0.9402509927749634,general_sample +139,aliskiren,0.1675613820552826,0.6590442657470703,general_sample +140,BRD-K93880770,0.16621443629264832,1.163874864578247,general_sample +141,BRD-K11989341,0.16608819365501404,0.6260548233985901,general_sample +142,BRD-K06714535,0.16518470644950867,1.6651201248168945,general_sample +143,BRD-K38588505,0.1649388074874878,0.26500213146209717,general_sample +144,BRD-K90705745,0.163993239402771,0.8484952449798584,general_sample +145,BRD-A41228941,0.16353529691696167,0.19634002447128296,general_sample +146,BRD-K04009734,0.16288873553276062,2.185487747192383,general_sample +147,AZD-6482,0.16243045032024384,1.668942928314209,general_sample +148,BRD-K30659453,0.1615477204322815,0.7147446274757385,general_sample +149,BRD-K50120786,0.161424919962883,2.1996657848358154,general_sample +150,BRD-K29635534,0.16122522950172424,0.5763481855392456,general_sample +151,BRD-K58072864,0.16102036833763123,1.7533352375030518,general_sample +152,pomalidomide,0.16026929020881653,1.1876976490020752,related_mechanism +153,BRD-K08482401,0.16006913781166077,1.2936341762542725,general_sample +154,BRD-K38320477,0.1577380746603012,2.0174200534820557,general_sample +155,BRD-K62768824,0.1574127972126007,0.2853902280330658,general_sample +156,BRD-K15672523,0.15602652728557587,0.25686198472976685,general_sample +157,BRD-K57064803,0.15426436066627502,2.142756223678589,general_sample +158,demeclocycline,0.15402083098888397,1.8052453994750977,general_sample +159,BRD-K81312543,0.15399408340454102,0.9155179858207703,general_sample +160,BRD-K65395273,0.15261712670326233,1.9891328811645508,general_sample +161,BRD-K91269653,0.15187959372997284,1.9947938919067383,general_sample +162,BRD-K98997045,0.15097758173942566,0.6952972412109375,general_sample +163,BRD-K44036769,0.14995774626731873,0.2860259413719177,general_sample +164,BRD-K69886508,0.14784830808639526,1.4674654006958008,general_sample +165,BRD-K01528688,0.1477918177843094,1.2374709844589233,general_sample +166,BRD-K45200389,0.14763033390045166,0.9359119534492493,general_sample +167,diphenhydramine,0.14630556106567383,1.6897326707839966,negative_control +168,BRD-K98710661,0.1454104781150818,0.5531234741210938,general_sample +169,BRD-K21635943,0.14415626227855682,1.7998664379119873,general_sample +170,sildenafil,0.14332857728004456,0.24974015355110168,related_mechanism +171,BRD-K14356681,0.14244668185710907,0.1876475214958191,general_sample +172,BRD-K46936109,0.1424178034067154,1.5194870233535767,general_sample +173,TH-302,0.14157085120677948,0.3173200488090515,general_sample +174,caffeine,0.14126121997833252,2.3276991844177246,negative_control +175,BRD-K53443165,0.14010101556777954,1.1625938415527344,general_sample +176,BRD-K73388776,0.139877051115036,-0.3698955178260803,general_sample +177,BRD-K64520484,0.1375540792942047,0.7576441764831543,general_sample +178,norethindrone,0.13644839823246002,1.541637659072876,negative_control +179,BRD-K13626871,0.13602501153945923,1.4089312553405762,general_sample +180,BRD-K72972600,0.13439995050430298,1.1684551239013672,general_sample +181,prednisolone,0.13360020518302917,0.8050864338874817,related_mechanism +182,BRD-K72232421,0.13345623016357422,2.274737596511841,general_sample +183,BRD-A43885598,0.1329154670238495,2.1677091121673584,general_sample +184,loperamide,0.13224346935749054,1.2183666229248047,negative_control +185,BRD-K09487323,0.13214272260665894,2.15812611579895,general_sample +186,BRD-K13307254,0.13211864233016968,1.0333926677703857,general_sample +187,BRD-K93541117,0.13047276437282562,2.168461322784424,general_sample +188,BRD-K40230514,0.12915246188640594,0.6537739038467407,general_sample +189,BRD-K07400331,0.12912264466285706,1.7710614204406738,general_sample +190,BRD-A78236793,0.12736168503761292,1.1413218975067139,general_sample +191,BRD-K95241801,0.12640994787216187,0.00014099478721618652,general_sample +192,BRD-K58023987,0.1262451410293579,0.5590131878852844,general_sample +193,BRD-K87669574,0.12599524855613708,1.7269186973571777,general_sample +194,BRD-K74553461,0.12537837028503418,2.184779644012451,general_sample +195,pipamperone,0.12477646768093109,0.1865198016166687,general_sample +196,JNJ-7706621,0.12447348982095718,0.35190898180007935,general_sample +197,BRD-K52183630,0.12413982301950455,2.428985118865967,general_sample +198,BRD-K57490754,0.12371955066919327,0.41090428829193115,general_sample +199,BRD-K72692763,0.12352791428565979,0.3076161742210388,general_sample +200,enilconazole,0.12276123464107513,1.0041255950927734,general_sample +201,BRD-K81524996,0.12215909361839294,0.6418701410293579,general_sample +202,BRD-K73155123,0.1192801296710968,0.6843189001083374,general_sample +203,BRD-K33528640,0.11878877878189087,0.5734832286834717,general_sample +204,BRD-K38181288,0.11808028817176819,1.7954599857330322,general_sample +205,QX-314,0.11777613312005997,2.060868501663208,general_sample +206,doxycycline,0.11709427833557129,1.192191243171692,negative_control +207,BRD-K68504618,0.11684578657150269,2.608350992202759,general_sample +208,levonorgestrel,0.11591199040412903,1.626094937324524,negative_control +209,itraconazole,0.11584744602441788,-0.09887504577636719,negative_control +210,BRD-K14483137,0.11480832099914551,1.6506483554840088,general_sample +211,BRD-K75990826,0.11396902799606323,0.9468180537223816,general_sample +212,VTP-27999,0.11371782422065735,0.8909273147583008,general_sample +213,BRD-K03235359,0.11317900568246841,2.4507906436920166,general_sample +214,diazepam,0.11256343871355057,1.7679884433746338,general_sample +215,simvastatin,0.11168932914733887,1.3542425632476807,related_mechanism +216,BRD-K38373457,0.11087027192115784,0.11042767763137817,general_sample +217,SB-258585,0.1105894148349762,0.9937275648117065,general_sample +218,azithromycin,0.11050024628639221,0.09130436182022095,negative_control +219,BRD-K97955841,0.11018981784582138,0.5305709838867188,general_sample +220,BRD-K97717522,0.10845574736595154,0.4835914969444275,general_sample +221,thalidomide,0.10836213827133179,2.0089573860168457,related_mechanism +222,reserpic-acid,0.10757994651794434,1.4694583415985107,general_sample +223,BRD-K51644197,0.10701952129602432,0.9039517641067505,general_sample +224,anastrozole,0.10358912497758865,0.9844931960105896,general_sample +225,BRD-K23961390,0.10208642482757568,-0.3048904538154602,general_sample +226,BRD-K82469533,0.10029269754886627,1.1784133911132812,general_sample +227,BRD-K68325500,0.0993058979511261,1.217383861541748,general_sample +228,verapamil,0.09837277233600616,1.61649489402771,general_sample +229,BRD-K00504156,0.09703031927347183,2.15120267868042,general_sample +230,BRD-A57107094,0.09696315228939056,2.0873100757598877,general_sample +231,BRD-K92333822,0.09657244384288788,0.3130739629268646,general_sample +232,BRD-K89675250,0.09552998840808868,1.096104621887207,general_sample +233,BRD-K20151898,0.09535171091556549,1.663360834121704,general_sample +234,BRD-K34811324,0.09506288915872574,1.636451005935669,general_sample +235,hydrocortisone,0.09483426809310913,1.3085203170776367,related_mechanism +236,BRD-K04232710,0.09364677965641022,0.9871090650558472,general_sample +237,BRD-K16539011,0.09267415851354599,2.0138421058654785,general_sample +238,lenalidomide,0.09157396852970123,1.7614586353302002,related_mechanism +239,dapivirine,0.08917832374572754,1.1204332113265991,general_sample +240,BRD-K82935485,0.0889597237110138,1.5348076820373535,general_sample +241,fluconazole,0.08861995488405228,0.7705867886543274,negative_control +242,miconazole,0.08847332000732422,0.3530352711677551,negative_control +243,BRD-K04643052,0.08728811889886856,1.8453972339630127,general_sample +244,trimethoprim,0.08714093267917633,1.5980050563812256,negative_control +245,BRD-K59887251,0.08359691500663757,-0.2983150780200958,general_sample +246,BRD-K97641878,0.08290493488311768,1.6386921405792236,general_sample +247,BRD-K36516410,0.08052213490009308,1.1813796758651733,general_sample +248,BRD-K63431240,0.08046489953994751,0.8348743915557861,general_sample +249,BRD-K07850148,0.0801025852560997,1.1315404176712036,general_sample +250,ketoconazole,0.07924821227788925,-0.2652745544910431,negative_control +251,BRD-K60994920,0.07906623929738998,1.9567651748657227,general_sample +252,medroxyprogesterone-acetate,0.07816928625106812,1.2908352613449097,negative_control +253,BRD-K42610174,0.07811297476291656,2.000000476837158,general_sample +254,BRD-K91339294,0.07743799686431885,1.6037771701812744,general_sample +255,BRD-K47885451,0.07666181027889252,1.7753000259399414,general_sample +256,BRD-K31042955,0.07525031268596649,-0.02010802924633026,general_sample +257,BRD-K85763971,0.07524567097425461,0.8542464971542358,general_sample +258,BRD-K73360774,0.07404312491416931,0.08928011357784271,general_sample +259,laropiprant,0.07321711629629135,1.2876667976379395,general_sample +260,BRD-K34462187,0.07217635214328766,2.635417938232422,general_sample +261,BRD-K00930335,0.07178701460361481,1.4565761089324951,general_sample +262,BRD-K98113345,0.07017907500267029,0.6807817220687866,general_sample +263,BRD-K39685946,0.06969957798719406,0.20744098722934723,general_sample +264,ketocholesterol,0.06865204125642776,1.517796516418457,general_sample +265,BRD-K54687541,0.06733538955450058,1.6019880771636963,general_sample +266,BRD-K89554882,0.06596627831459045,0.9116415977478027,general_sample +267,adrenosterone,0.06559927761554718,1.6499176025390625,general_sample +268,oseltamivir-carboxylate,0.06034604460000992,1.8670696020126343,general_sample +269,BRD-K33514849,0.06033096835017204,0.9989179968833923,general_sample +270,BRD-K62265198,0.0597817488014698,2.0923120975494385,general_sample +271,BRD-K01520595,0.059522368013858795,1.1388102769851685,general_sample +272,BRD-K75965029,0.05909409373998642,0.8850012421607971,general_sample +273,BRD-K60379529,0.056600529700517654,1.3942689895629883,general_sample +274,BRD-K77323800,0.05628868564963341,0.76136714220047,general_sample +275,BRD-K93200353,0.05566142499446869,1.367753028869629,general_sample +276,BRD-K39337865,0.0547766387462616,1.8628408908843994,general_sample +277,BRD-K34412442,0.05267806351184845,1.960115909576416,general_sample +278,BRD-K41866979,0.05012091249227524,0.5144319534301758,general_sample +279,ticagrelor,0.049006469547748566,0.0014074146747589111,related_mechanism +280,BRD-K40829514,0.04817928373813629,0.4225776195526123,general_sample +281,BRD-K02257023,0.046159256249666214,1.8649888038635254,general_sample +282,BRD-K47105324,0.042336881160736084,0.6837743520736694,general_sample +283,BRD-K79992300,0.04170645773410797,1.2891095876693726,general_sample +284,BRD-K43905345,0.040275897830724716,2.065195322036743,general_sample +285,BRD-K42269438,0.03933650255203247,1.3552615642547607,general_sample +286,BRD-K56349799,0.03727919980883598,1.6540254354476929,general_sample +287,dexamethasone-acetate,0.03663018345832825,1.0805904865264893,general_sample +288,BRD-K27675859,0.03387710824608803,1.9398986101150513,general_sample +289,SB-525334,0.032205577939748764,1.5904217958450317,general_sample +290,rucaparib,0.03211560472846031,0.8603106141090393,general_sample +291,BRD-K45787391,0.031139448285102844,0.741227388381958,general_sample +292,pregnenolone,0.030906977131962776,1.5815171003341675,general_sample +293,BRD-K19645992,0.03016829863190651,1.7404652833938599,general_sample +294,BRD-K50309272,0.028783636167645454,1.5358946323394775,general_sample +295,dexamethasone,0.027089979499578476,0.6470061540603638,related_mechanism +296,torin-1,0.02660956233739853,-0.252296507358551,general_sample +297,BRD-K08246588,0.02257409133017063,2.5512869358062744,general_sample +298,BRD-K87400432,0.02154897153377533,1.3216978311538696,general_sample +299,AV-608,0.020470259711146355,1.0707037448883057,general_sample diff --git a/docs/structure_binding_notes.md b/docs/structure_binding_notes.md index b1e2d59..1b8ba22 100644 --- a/docs/structure_binding_notes.md +++ b/docs/structure_binding_notes.md @@ -160,8 +160,35 @@ model. The screen mishandles non-standard chemotypes (prodrugs, macrocycles). The screening pipeline is validated. Next: run the full set (incl. the 240 random + negatives) to hunt for NON-obvious HDAC2 binders (the actual discovery run), ~$15-20. +## Step 7 — Full 300-drug discovery screen vs HDAC2 (2026-06-26) + +`modal run gpu/modal_app.py::screen` — 299 drugs co-folded vs HDAC2 (+Zn), ranked by P(binder). +Corrected pipeline: 1 MSA query (computed once, reused), 299/299 screened, 0 failures, ~$5-8. + +**Scale validation:** HDAC inhibitors occupy ranks 1-9 (trichostatin-A, vorinostat, panobinostat, +belinostat, scriptaid, mocetinostat, entinostat, apicidin; all ≥0.99), weak valproic-acid demoted +to 0.90. The pilot held at 300. + +**DECISIVE — specificity (the thing connectivity could NOT do):** the best-scoring negative +control is cetirizine at **rank 44, P=0.39**. All 26 negative controls (antifungals, +antihistamines, antibiotics, hormones) rank low — co-folding REJECTS the unrelated drugs. This is +the exact failure mode that sank the connectivity approach (negative controls ranked top there); +structure-based binding has the specificity expression-connectivity fundamentally lacked. + +**Discovery hits (non-obvious high-P binders):** BG-1003 (rank 5, P=0.997, general_sample), +BRD-K14666757 (10, 0.968), JW55 (11, 0.936, a tankyrase inhibitor), FIT (13, 0.831). 11 drugs +score P>0.9 = 8 known HDAC inhibitors + 3 non-obvious. BG-1003 is the standout — a "random" LINCS +compound scoring as a near-certain HDAC2 binder above several known inhibitors. + +**Honest caveats:** BG-1003 may be a known HDAC inhibitor that landed in the random sample (→ +validation, not novelty) — needs an identity/literature check before any claim. Several hits are +unannotated BRD tool compounds. Binding != efficacy. The screen still mishandles prodrugs/ +macrocycles (romidepsin-type false negatives). Full ranking: docs/results/screen_HDAC2_full.csv. + ## Next steps -- [ ] Full screen (300 drugs) vs HDAC2 — discovery run for non-obvious binders. +- [ ] Identity-check BG-1003 and the BRD hits (ChEMBL/literature): known HDAC binders or novel? +- [ ] Pose-RMSD the top non-obvious hits (geometry sanity, like vorinostat). +- [ ] Extend the screen to other validated HbF/hemoglobin targets; integrate with the expression layer. - [ ] Investigate PKR: allosteric site may need the full assembly / better pocket definition. - [ ] Phase 2 screen: rank the ~300-drug set against HDAC2 (the validated target) by P(binder); positive-control recovery test at screen scale.