Machine learning discoveries of FANCD2-X synergy in ETC-1922159 treated colorectal cancer cells

Machine learning discoveries of FANCD2-X synergy in ETC-1922159 treated colorectal cancer cells shriprakash sinha Independent Researcher; Orcid ID : orcid.org/0000-0001-7027-5788 104-Madhurisha Heights Phase 1, Risali, Bhilai-490006, India Abstract Fanconi anemia complementation group D2 (FANCD2) is one of the members of FA complementation group, that is activated in response to DNA damage and is involved in DNA repair and regulation of genomic stability. FANCD2 works with other members of FANC group along with other proteins/genes, to carry out its required functionality. Further, disruption of FA/BRCA pathway has been implicated in cancer progression. In colorectal cancer (CRC) cells treated with ETC-1922159, FANCD2 was found to be down regulated along with other genes. A recently developed search engine ranked combinations of FANCD2-X (X, a particular gene/protein) at 2nd order level after drug administration. Some of these combinations have been tested and established in wet lab, however many have been pointed out by the search engine that are yet to be explored/tested. These rankings reveal which FANCD2-X combinations might be working synergistically in CRC. In this research work, I cover combinations of FANCD2 with, REV3 like DNA directed polymerase zeta catalytic subunit (REV3L), Bloom syndrome RecQ like helicase (BLM), MRE11 homolog double strand break repair nuclease (MRE11A), Wnt family member 10B (WNT10B), ubiquitin like with PHD and ring finger domains 1 (UHRF1), hes family bHLH transcription factor 1 (HES1), BRCA DNA repair associated (BRCA), RAD51 recombinase (RAD51), ERCC excision repair endonuclease non-catalytic subunit (ERCC), KIAA, X-ray repair cross complementing (XRCC), structural maintenance of chromosomes (SMC), WD repeat domain containing (WDR), ubiquitin conjugating enzyme E2 (UBE2), cell division cycle (CDC), importin (IPO), aldehyde dehydrogenase family member (ALDH), H2A variant histone (H2A), heat shock protein (HSP), cyclin dependent kinase (CDK), dynein axonemal heavy chain (DNAH), forkhead box (FOX), ring finger protein (RNF), E2F transcription factor (E2F) and small nucleolar RNA host gene (SNHG) family. Keywords: FANCD2, Porcupine inhibitor ETC-1922159, Sensitivity analysis, Colorectal cancer. IML dicoveries of FANCD2-X synergy in ETC-1922159 treated CRC cells Email address: [email protected] (shriprakash sinha) 1Aspects of unpublished work were presented in a poster session at the first Wnt Gordon Research Conference, from 6-11 August 2017, held in Stowe, VT 05672, USA. Preprint submitted to Preprint January 4, 2025 1. Introduction 1.1. Fanconi anemia Fanconi anemia (FA) is an autosomal recessive genetic disease that causes impaired response to DNA damage in the FA/BRCA pathway. FA was first described in 1927 by Guido Fanconi. D’Andrea [1] give a perspective on the role of FA complementation group proteins that are encoded by FANC genes. Briefly, D’Andrea [1] sumarizes the basic working background of the FA/BRCA pathway as follows - The FA proteins (A, C, E, F, G) are assembled in a nuclear protein complex along with BLM as a subunit. This complex is then required for the monoubiquitination of the downstream FANCD2. Next, monoubiquitinated FANCD2 assembles with the breast cancer susceptibility proteins, BRCA1 and BRCA2. The nuclear foci also contain the RAD51 and the MRE11/RAD50/NBS1 (M/R/N) complex. Disruption of genes in this pathway results in cellular hypersensitivity to DNA cross-linking agents, chromosome instability, and a propensity to cancer progression. 1.2. FANCD2 Houghtaling et al. [2] via FANCD2 knockout experiments in mice, showed that cellular sensitivity to DNA interstrand cross-links, perinatal lethality, germ cell loss, epithelial cancers and microphthalmia. Their work provides the first molecular link between the FA pathway and epithelial cancers. van Twest et al. [3] provide a detailed biochemical analysis and mechanistic basis of FANCI and FANCD2 monoubiquitination by the FA complex. They show that FANCB dimer coordinates the FANCD2 and FANCI monoubiquitination by two FANCL RING-ligases. FANCD2 was found to be down regulated in colorectal cancer cell lines after the treatment of ETC-1922159 drug, along with other genes, as observed in Madan et al. [4]. Many of the combinations of FANCD2 with other genes have been tested, established and reported in wet lab experiments, however there remain a range of combinations that are yet to be explored and tested. It would be nice to observe if there is any connection between the independently observed factors in the form of unknown biological hypotheses. To solve the issue, the next section discusses a solution to the problem. 1.3. Combinatorial search problem and a possible solution In a recently published work Sinha [5], a frame work of a search engine was developed which can rank combinations of factors (genes/proteins) in a signaling pathway. Readers are requested to go through the adaptation of the above mentioned work for gaining deeper insight into the working of the pipeline and its use of published data set generated after administration of ETC-1922159, Sinha [6]. The work uses SVM package by Joachims [7] in https://www.cs.cornell.edu/people/tj/svm_light/svm_ rank.html. I use the adaptation to rank 2nd order gene combinations. 2 2. Results & Discussion 2.1. FANCD2 related synergies 2.1.1. FANCD2 - REV3L / BLM / MRE11A / WNT10B / UHRF1 / HES1 Bhat et al. [8] observe that experimental depletion of REV3 leads to increase in anaphase bridges, common fragile site (CFS) expression and chromosomal breaks/gaps. The genomic instability induced by REV3 depletion results in increased metaphase-specific FANCD2 foci formation and FANCD2 positive anaphase bridges, thus showing a connection between REV3 and FANCD2. Pichierri et al. [9] show that BLM and FANCD2 colocalise and co-immunoprecipitate in response to crosslinked DNA and stalled replication forks. Also, BLM and the FA core complex are both necessary for the assembly of the MRE11 complex. Further, the connection between MRE11 and FANCD2 is established as Roques et al. [10] show that inhibition of MRE11, NBS1 or RAD50 leads to a destabilization of FANCD2. Kaur et al. [11] found that FANCD2 was dependent on WNT/β-catenin signaling in WNT-high cancers, and treatment with a PORCN inhibitor creates a BRCA-like state. FA pathway is critical for the response to toxic DNA interstrand crosslinks (ICLs). Liang et al. [12] showed that reduction of cellular levels of UHRF1 by RNAi attenuates the FA pathway and knockdown cells cause a reduction in FANCD2 foci formation. They observed that UHRF1 is required for the recruitment of FANCD2 to ICLs, which allows the DNA repair process to initiate. Finally, Tremblay et al. [13] show that HES1 is an interactor of FA core complex and is required for FANCD2 monoubiquitination. Taken together, all these experiments show a direct connection/synergy of the involved members with FANCD2. In colorectal cancer cells treated with ETC-1922159, these individual members (or their variants) and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of individial members and FANCD2, that were down regulated. Table 1 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 2 generated from analysis of the ranks in table 1. The table 1 shows rankings of individual members w.r.t FANCD2. BLM - FANCD2 shows low ranking of 184 (laplace) and 305 (rbf). WNT10B - FANCD2 shows low ranking of 365 (laplace) and 307 (rbf). UHRF1 - FANCD2 shows low ranking of 1356 (laplace), 880 (linear) and 1268 (rbf). HES6 - FANCD2 shows low ranking of 1495 (laplace), 1182 (linear) and 1192 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, REV3L and MRE11A showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. One can also interpret the results of the table 1 graphically, with the following influences - •INDIVIDUAL family w.r.t FANCD2 with FANCD2 −>BLM, FANCD2 −>WNT10B, FANCD2 −>UHRF1, and FANCD2 −>HES6; . 2.1.2. FANCD2 - BRCA DNA damage activates FA pathway, leading to monoubiquitination of FANCD2 protein and targeting nuclear foci containing BRCA1. Taniguchi et al. [14] show that this 3 RANKING INDIVIDUAL MEMBERS VS FANCD2 RANKING OF INDIVIDUAL MEMBERS W.R.TFANCD2 laplace linear rbf REV3L - FANCD2 1872 1049 2677 BLM - FANCD2 184 2738 305 MRE11A - FANCD2 2042 1744 1344 WNT10B - FANCD2 365 1891 307 UHRF1 - FANCD2 1356 880 1268 HES6 - FANCD2 1495 1182 1192 Table 1: 2nd order interaction ranking between FANCD2 VS INDIVIDUAL family members. UNEXPLORED COMBINATORIAL HYPOTHESES INDIVIDUAL members w.r.t FANCD2 BLM / WNT10B / UHRF1 / HES6 FANCD2 Table 2: 2nd order combinatorial hypotheses between FANCD2 and INDIVIDUAL family members. monoubiquitinated FANCD2 colocalizes with BRCA1 and RAD51 in S-phasespecific nuclear foci. Similar interaction was reported by Garcia-Higuera et al. [15]. Roy et al. [16] discuss that both BRCA1 and BRCA2 work in a common pathway, however they work at different stages in DNA damage response (DDR). BRCA1 functions in both checkpoint activation and DNA repair, while BRCA2 is a mediator of homologous recombination. In colorectal cancer cells treated with ETC-1922159, these BRCA family members and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of BRCA members and FANCD2, that were down regulated. Table 3 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 4 generated from analysis of the ranks in table 3. The table 3 shows rankings of individual members w.r.t FANCD2. BRCA2 - FANCD2 shows low ranking of 354 (laplace) and 377 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, BRCA1 showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. One can also interpret the results of the table 3 graphically, with the following influences - •BRCA family w.r.t FANCD2 with FANCD2 −>BRCA2. 4 RANKING BRCA MEMBERS VS FANCD2 RANKING OF BRCA MEMBERS W.R.TFANCD2 laplace linear rbf BRCA2 - FANCD2 354 2668 377 BRCA1 - FANCD2 1772 244 2238 Table 3: 2nd order interaction ranking between FANCD2 VS BRCA family members. UNEXPLORED COMBINATORIAL HYPOTHESES BRCA members w.r.t FANCD2 BRCA2 FANCD2 Table 4: 2nd order combinatorial hypotheses between FANCD2 and BRCA family members. 2.1.3. FANCD2 - RAD51 Based on the previous section where it was cited that Taniguchi et al. [14] show that the monoubiquitinated FANCD2 colocalizes with BRCA1 and RAD51 in S-phasespecific nuclear foci, it can be assumed that there is a synergy between RAD51 and FANCD2. In colorectal cancer cells treated with ETC-1922159, these RAD51 family members and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of RAD51 members and FANCD2, that were down regulated. Table 5 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 6 generated from analysis of the ranks in table 5. The table 5 shows rankings of individual members w.r.t FANCD2. RAD51 - FANCD2 shows low ranking of 885 (laplace) and 1383 (rbf). RAD51AP1 - FANCD2 shows low ranking of 644 (linear) and 1291 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, RAD51C showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. One can also interpret the results of the table 5 graphically, with the following influences - •RAD51 family w.r.t FANCD2 with FANCD2 −>RAD-51/51AP1. 2.1.4. FANCD2 - ERCC ERCC1 is identified as a participant in nucleotide excision repair and has a possible relationship with FA pathway. McCabe et al. [17] show that the lack of FANCD2 focus formation is greater with depletion of ERCC1 than the lack of monoubiquiti5 RANKING RAD51 MEMBERS VS FANCD2 RANKING OF RAD51 MEMBERS W.R.TFANCD2 laplace linear rbf RAD51 - FANCD2 885 1995 1383 RAD51AP1 - FANCD2 1734 644 1291 RAD51C - FANCD2 54 2399 2566 Table 5: 2nd order interaction ranking between FANCD2 VS RAD51 family members. UNEXPLORED COMBINATORIAL HYPOTHESES RAD51 members w.r.t FANCD2 RAD-51/51AP1 FANCD2 Table 6: 2nd order combinatorial hypotheses between FANCD2 and RAD51 family members. nated FANCD2 formation. This suggests a relation between ERCC1 and FANCD2 and in particular FA pathway. In colorectal cancer cells treated with ETC-1922159, these ERCC family members and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of ERCC members and FANCD2, that were down regulated. Table 7 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 8 generated from analysis of the ranks in table 7. The table 7 shows rankings of individual members w.r.t FANCD2. ERCC8 - FANCD2 shows low ranking of 371 (laplace) and 997 (rbf). ERCC6L - FANCD2 shows low ranking of 1209 (laplace) and 520 (linear). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. RANKING ERCC MEMBERS VS FANCD2 RANKING OF ERCC MEMBERS W.R.TFANCD2 laplace linear rbf ERCC8 - FANCD2 371 1951 997 ERCC6L - FANCD2 1209 520 1884 Table 7: 2nd order interaction ranking between FANCD2 VS ERCC family members. 6 One can also interpret the results of the table 7 graphically, with the following influences - •ERCC family w.r.t FANCD2 with FANCD2 −>ERCC-8/6L. UNEXPLORED COMBINATORIAL HYPOTHESES ERCC members w.r.t FANCD2 ERCC-8/6L FANCD2 Table 8: 2nd order combinatorial hypotheses between FANCD2 and ERCC family members. 2.1.5. FANCD2 - KIAA Kikuno et al. [18] document a range of over 2000 novel human genes that are designated KIAA plus a four-digit number. MacKay et al. [19] describe a highly conserved protein, KIAA1018 that interacts with, and is recruited to sites of DNA damage by, the monoubiquitinated form of FANCD2. In colorectal cancer cells treated with ETC1922159, these KIAA family members and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of KIAA members and FANCD2, that were down regulated. Table 9 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 10 generated from analysis of the ranks in table 9. The table 9 shows rankings of individual members w.r.t FANCD2. KIAA0586 - FANCD2 shows low ranking of 211 (laplace), 703 (linear) and 17 (rbf). KIAA1524 - FANCD2 shows low ranking of 225 (laplace) and 481 (rbf). KIAA1257 - FANCD2 shows low ranking of 749 (laplace), 1173 (linear) and 587 (rbf). KIAA0020 - FANCD2 shows low ranking of 979 (laplace) and 1332 (rbf). KIAA1324 - FANCD2 shows low ranking of 1466 (laplace), 413 (linear) and 1410 (rbf). KIAA1731 - FANCD2 shows low ranking of 1549 (laplace) and 1592 (linear). KIAA1586 - FANCD2 shows low ranking of 1233 (linear) and 277 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, KIAA0101, KIAA1143, KIAA1244 and KIAA1430 showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. One can also interpret the results of the table 9 graphically, with the following influences - •KIAA family w.r.t FANCD2 with FANCD2 −>KIAA-0586 / 1524 / 1257 / 0020 / 1324 / 1731 / 1586. 2.1.6. FANCD2 - XRCC XRCC2 is one of five somatic RAD51 paralogs and Andreassen and Hanenberg [20] state it appears to function downstream in the FA pathway, as it is not required for FANCD2 monoubiquitination. In colorectal cancer cells treated with ETC-1922159, these XRCC family members and FANCD2, were found to be down regulated and 7 RANKING KIAA MEMBERS VS FANCD2 RANKING OF KIAA MEMBERS W.R.TFANCD2 laplace linear rbf KIAA0586 - FANCD2 211 703 17 KIAA1524 - FANCD2 225 2578 481 KIAA1257 - FANCD2 749 1173 587 KIAA0020 - FANCD2 979 2291 1332 KIAA1324 - FANCD2 1466 413 1410 KIAA1731 - FANCD2 1549 1592 2223 KIAA1586 - FANCD2 1667 1233 277 KIAA0101 - FANCD2 2118 570 2103 KIAA1143 - FANCD2 2279 1094 2043 KIAA1244 - FANCD2 2341 777 2453 KIAA1430 - FANCD2 2468 1138 2075 Table 9: 2nd order interaction ranking between FANCD2 VS KIAA family members. UNEXPLORED COMBINATORIAL HYPOTHESES KIAA members w.r.t FANCD2 KIAA-0586/1524/1257/0020/1324/1731/1586 FANCD2 Table 10: 2nd order combinatorial hypotheses between FANCD2 and KIAA family members. recorded independently. I was able to rank 2nd order combinations of XRCC members and FANCD2, that were down regulated. Table 11 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 12 generated from analysis of the ranks in table 11. The table 11 shows rankings of individual members w.r.t FANCD2. XRCC6BP1 - FANCD2 shows low ranking of 1545 (linear) and 1200 (rbf). XRCC2 - FANCD2 shows low ranking of 111 (linear) and 1269 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, XRCC1 and XRCC6 showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treat8 ment. RANKING XRCC MEMBERS VS FANCD2 RANKING OF XRCC MEMBERS W.R.TFANCD2 laplace linear rbf XRCC6BP1 - FANCD2 1720 1545 1200 XRCC2 - FANCD2 2002 111 1269 XRCC1 - FANCD2 2096 1096 2170 XRCC6 - FANCD2 2694 215 2539 Table 11: 2nd order interaction ranking between FANCD2 VS XRCC family members. One can also interpret the results of the table 11 graphically, with the following influences - •XRCC family w.r.t FANCD2 with FANCD2 −>XRCC-6BP1/2. UNEXPLORED COMBINATORIAL HYPOTHESES XRCC members w.r.t FANCD2 XRCC-6BP1/2 FANCD2 Table 12: 2nd order combinatorial hypotheses between FANCD2 and XRCC family members. 2.1.7. FANCD2 - SMC Rossi et al. [21] show that SMC5 functions downstream of FANCD2 ubiquitylation with FANCC and SMC5/6 acts jointly with FANCD2 to mediate DNA repair and prevent genomic instability in human cells. Thus there is synergy between SMC and FANCD2. In colorectal cancer cells treated with ETC-1922159, these SMC family members and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of SMC members and FANCD2, that were down regulated. Table 13 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 14 generated from analysis of the ranks in table 13. The table 13 shows rankings of individual members w.r.t FANCD2. SMC1A - FANCD2 shows low ranking of 1156 (laplace), 88 (linear) and 1502 (rbf). SMC4 - FANCD2 shows low ranking of 44 (laplace) and 83 (rbf). SMC2 - FANCD2 shows low ranking of 229 (laplace) and 236 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. 9 UNEXPLORED COMBINATORIAL HYPOTHESES ALDH members w.r.t FANCD2 ALDH-3A1/5A1/7A1/1B1/9A1 FANCD2 Table 24: 2nd order combinatorial hypotheses between FANCD2 and ALDH family members. and recorded independently. I was able to rank 2nd order combinations of H2A members and FANCD2, that were down regulated. Table 25 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 26 generated from analysis of the ranks in table 25. The table 25 shows rankings of individual members w.r.t FANCD2. H2AFX (or H2AX) - FANCD2 shows low ranking of 593 (laplace) and 324 (linear). H2AFV - FANCD2 shows low ranking of 791 (laplace), 1329 (linear). H2AFZ - FANCD2 shows low ranking of 1292 (laplace) and 634 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. RANKING H2A MEMBERS VS FANCD2 RANKING OF H2A MEMBERS W.R.TFANCD2 laplace linear rbf H2AFX (or H2AX) - FANCD2 593 324 2008 H2AFV - FANCD2 791 1329 1659 H2AFZ - FANCD2 1292 1921 634 Table 25: 2nd order interaction ranking between FANCD2 VS H2A family members. One can also interpret the results of the table 25 graphically, with the following influences - •H2A family w.r.t FANCD2 with FANCD2 −>H2A-FX/FV/FZ. UNEXPLORED COMBINATORIAL HYPOTHESES H2A members w.r.t FANCD2 H2A-FX/FV/FZ FANCD2 Table 26: 2nd order combinatorial hypotheses between FANCD2 and H2A family members. 16 2.1.14. FANCD2 - HSP Oda et al. [28] showed that HSP90 associated with FANCA, and disruption of this association by treatment with 17-AAG induces proteasomal degradation and cytoplasmic relocalization of FANCA, which lead to impaired activation of FANCD2. Thus there is an established connection between HSP90 and FANCD2. In colorectal cancer cells treated with ETC-1922159, these HSP family members and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of HSP members and FANCD2, that were down regulated. Table 27 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 28 generated from analysis of the ranks in table 27. The table 27 shows rankings of individual members w.r.t FANCD2. HSPE1 - FANCD2 shows low ranking of 336 (laplace), 1246 (linear) and 745 (rbf). HSPD1 - FANCD2 shows low ranking of 385 (laplace), 388 (linear) and 617 (rbf). HSPB6 - FANCD2 shows low ranking of 1000 (laplace) and 520 (rbf). HSPA9 - FANCD2 shows low ranking of 370 (linear) and 1514 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, HSPA4L and HSP4A showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. RANKING HSP MEMBERS VS FANCD2 RANKING OF HSP MEMBERS W.R.TFANCD2 laplace linear rbf laplace linear rbf HSPE1 - FANCD2 336 1246 745 HSPD1 - FANCD2 385 388 617 HSPA4L - FANCD2 520 2211 2202 HSPB6 - FANCD2 1000 2421 520 HSPA9 - FANCD2 2078 370 1514 HSPA4 - FANCD2 2621 25 2650 Table 27: 2nd order interaction ranking between FANCD2 VS HSP family members. One can also interpret the results of the table 27 graphically, with the following influences - •HSP family w.r.t FANCD2 with FANCD2 −>HSP-E1/D1/B6/A9. UNEXPLORED COMBINATORIAL HYPOTHESES HSP members w.r.t FANCD2 HSP-E1/D1/B6/A9 FANCD2 Table 28: 2nd order combinatorial hypotheses between FANCD2 and HSP family members. 2.1.15. FANCD2 - CDK Blazek et al. [29] show that a variant of cyclin-CDK complex namely CYCK/CDK12 regulated expression of a small subset of human genes. Depletion of CYCK/CDK12 17 resulted in decreased expression of BRCA1, FANCD2, ATR and FANCI. Via this regulation, it is thought that the CYCK/CDK12 protects the cells from genomic instability. In colorectal cancer cells treated with ETC-1922159, these CDK family member and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of CDK members and FANCD2, that were down regulated. Table 29 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 30 generated from analysis of the ranks in table 29. The table 29 shows rankings of individual members w.r.t FANCD2. CDK5RAP1 - FANCD2 shows low ranking of 1305 (laplace), 566 (linear) and 1196 (rbf). CDK5RAP2 - FANCD2 shows low ranking of 34 (linear) and 921 (rbf). CDK6 - FANCD2 shows low ranking of 337 (linear) and 1409 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, CDK4 showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. RANKING CDK MEMBERS VS FANCD2 RANKING OF CDK MEMBERS W.R.TFANCD2 laplace linear rbf CDK4 - FANCD2 918 1632 1890 CDK5RAP1 - FANCD2 1305 566 1196 CDK5RAP2 - FANCD2 2064 34 921 CDK6 - FANCD2 2463 337 1409 Table 29: 2nd order interaction ranking between FANCD2 VS CDK family members. One can also interpret the results of the table 29 graphically, with the following influences - •CDK family w.r.t FANCD2 with FANCD2 −>CDK-6/5RAP1/5RAP2. UNEXPLORED COMBINATORIAL HYPOTHESES CDK members w.r.t FANCD2 CDK-6/5RAP1/5RAP2 FANCD2 Table 30: 2nd order combinatorial hypotheses between FANCD2 and CDK family members. 18 2.1.16. FANCD2 - DNAH Chang et al. [30] show involvement of DNAH2 in the regulation of ubiquitination and nuclear localization of FANCD2 upon the DNA damage. In colorectal cancer cells treated with ETC-1922159, these DNAH family member and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of DNAH members and FANCD2, that were down regulated. Table 31 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 32 generated from analysis of the ranks in table 31. The table 31 shows rankings of individual members w.r.t FANCD2. DNAH11 - FANCD2 shows low ranking of 762 (laplace), 1463 (linear) and 567 (rbf). DNAH14 - FANCD2 shows low ranking of 1252 (laplace) and 229 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. RANKING DNAH MEMBERS VS FANCD2 RANKING OF DNAH MEMBERS W.R.TFANCD2 laplace linear rbf DNAH11 - FANCD2 762 1463 567 DNAH14 - FANCD2 1252 1703 229 Table 31: 2nd order interaction ranking between FANCD2 VS DNAH family members. One can also interpret the results of the table 31 graphically, with the following influences - •DNAH family w.r.t FANCD2 with FANCD2 −>DNAH-11/14. UNEXPLORED COMBINATORIAL HYPOTHESES DNAH members w.r.t FANCD2 DNAH-11/14 FANCD2 Table 32: 2nd order combinatorial hypotheses between FANCD2 and DNAH family members. 2.1.17. FANCD2 - FOX Li et al. [31] report the functional interaction of FANCD2 and the forkhead transcription factor forkhead box (FOX) O3a i.e FOXO3A. FOXO3A formed a complex with FANCD2 foci in response to oxidative stress. In colorectal cancer cells treated with ETC-1922159, these FOX family member and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of FOX members and FANCD2, that were down regulated. 19 Table 33 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 34 generated from analysis of the ranks in table 33. The table 33 shows rankings of individual members w.r.t FANCD2. FOXJ1 - FANCD2 shows low ranking of 830 (laplace) and 972 (rbf). FOXD2-AS1 - FANCD2 shows low ranking of 1154 (laplace) and 611 (linear). FOXM1 - FANCD2 shows low ranking of 44 (linear) and 1142 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, FOXA2 showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. RANKING FOX MEMBERS VS FANCD2 RANKING OF FOX MEMBERS W.R.TFANCD2 laplace linear rbf FOXJ1 - FANCD2 830 2003 972 FOXD2-AS1 - FANCD2 1154 611 2023 FOXM1 - FANCD2 1868 44 1142 FOXA2 - FANCD2 2101 1184 2240 Table 33: 2nd order interaction ranking between FANCD2 VS FOX family members. One can also interpret the results of the table 33 graphically, with the following influences - •FOX family w.r.t FANCD2 with FANCD2 −>FOX-J1/D2-AS1/M1. UNEXPLORED COMBINATORIAL HYPOTHESES FOX members w.r.t FANCD2 FOX-J1/D2-AS1/M1 FANCD2 Table 34: 2nd order combinatorial hypotheses between FANCD2 and FOX family members. 2.1.18. FANCD2 - RNF Yan et al. [32] show that FAAP20 binds the ubiquitin product of RNF8-UBC13. RNF8FAAP20 cascade is needed for efficient FANCD2 monoubiquitination. In colorectal cancer cells treated with ETC-1922159, these RNF family member and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of RNF members and FANCD2, that were down regulated. Table 35 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 36 generated from analysis of the ranks in table 35. 20 The table 35 shows rankings of individual members w.r.t FANCD2. RNF157-AS1 - FANCD2 shows low ranking of 399 (laplace) and 533 (linear). RNF166 - FANCD2 shows low ranking of 659 (laplace) and 526 (rbf). RNF20 - FANCD2 shows low ranking of 1227 (laplace), 247 (linear) and 1331 (rbf). RNF26 - FANCD2 shows low ranking of 1476 (laplace), 873 (linear) and 1540 (rbf). RNF144B - FANCD2 shows low ranking of 1504 (laplace) and 1095 (rbf). RNF157 - FANCD2 shows low ranking of 1553 (laplace), 417 (linear) and 449 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, RNF220 and RNF43 showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. RANKING RNF MEMBERS VS FANCD2 RANKING OF RNF MEMBERS W.R.TFANCD2 laplace linear rbf RNF157-AS1 - FANCD2 399 533 1742 RNF166 - FANCD2 659 1764 526 RNF20 - FANCD2 1227 247 1331 RNF43 - FANCD2 1366 1571 2048 RNF26 - FANCD2 1476 873 1540 RNF144B - FANCD2 1504 1733 1095 RNF157 - FANCD2 1553 417 449 RNF220 - FANCD2 2716 14 2668 Table 35: 2nd order interaction ranking between FANCD2 VS RNF family members. One can also interpret the results of the table 35 graphically, with the following influences - •RNF family w.r.t FANCD2 with FANCD2 −>RNF-157-AS1/166/20/26/144B/157. UNEXPLORED COMBINATORIAL HYPOTHESES RNF members w.r.t FANCD2 RNF-157-AS1/166/20/26/144B/157 FANCD2 Table 36: 2nd order combinatorial hypotheses between FANCD2 and RNF family members. 21 2.1.19. FANCD2 - E2F Mitxelena et al. [33] showed that E2F7 modulates the DNA repair program as knockdown experiment of E2F7 lead to a reduction in 53BP1 and FANCD2 foci. In colorectal cancer cells treated with ETC-1922159, these E2F family member and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of E2F members and FANCD2, that were down regulated. Table 37 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 38 generated from analysis of the ranks in table 37. The table 37 shows rankings of individual members w.r.t FANCD2. E2F1 - FANCD2 shows low ranking of 200 (laplace) and 469 (rbf). E2F2 - FANCD2 shows low ranking of 472 (laplace) and 140 (rbf). E2F7 - FANCD2 shows low ranking of 482 (laplace) and 985 (rbf). E2F8 - FANCD2 shows low ranking of 192 (linear) and 582 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, E2F5 showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. RANKING E2F MEMBERS VS FANCD2 RANKING OF E2F MEMBERS W.R.TFANCD2 laplace linear rbf E2F1 - FANCD2 200 1944 469 E2F2 - FANCD2 472 2345 140 E2F7 - FANCD2 482 2500 985 E2F8 - FANCD2 1873 192 582 E2F5 - FANCD2 2537 435 2079 Table 37: 2nd order interaction ranking between FANCD2 VS E2F family members. One can also interpret the results of the table 37 graphically, with the following influences - •E2F family w.r.t FANCD2 with FANCD2 −>E2F-1/2/7/8. UNEXPLORED COMBINATORIAL HYPOTHESES E2F members w.r.t FANCD2 E2F-1/2/7/8 FANCD2 Table 38: 2nd order combinatorial hypotheses between FANCD2 and E2F family members. 22 2.1.20. FANCD2 - SNHG In hepatocellular carcinoma, [34] showed that SNHG1 upregulated the expression of FANCD2 and G6PD, thus inhibiting ferroptosis. In colorectal cancer cells treated with ETC-1922159, these SNHG family member and FANCD2, were found to be down regulated and recorded independently. I was able to rank 2nd order combinations of SNHG members and FANCD2, that were down regulated. Table 39 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 40 generated from analysis of the ranks in table 39. The table 39 shows rankings of individual members w.r.t FANCD2. SNHG3 - FANCD2 shows low ranking of 339 (laplace) and 1259 (linear) SNHG6 - FANCD2 shows low ranking of 551 (laplace) and 74 (rbf). SNHG10 - FANCD2 shows low ranking of 866 (laplace) and 1442 (rbf). SNHG5 - FANCD2 shows low ranking of 1140 (laplace) and 674 (rbf). SNHG7 - FANCD2 shows low ranking of 1220 (laplace) and 694 (linear) SNHG17 - FANCD2 shows low ranking of 1524 (laplace), 454 (linear) and 1198 (rbf). SNHG15 - FANCD2 shows low ranking of 1464 (linear) and 322 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, SNHG16, SNHG1, SNHG18 and SNHG8 showed high ranking with FANCD2, thus indicating that they might not be working synergistically with FANCD2, before the drug treatment. RANKING SNHG MEMBERS VS FANCD2 RANKING OF SNHG MEMBERS W.R.TFANCD2 laplace linear rbf laplace linear rbf SNHG3 - FANCD2 339 1259 1759 SNHG6 - FANCD2 551 1943 74 SNHG10 - FANCD2 866 1960 1442 SNHG5 - FANCD2 1140 1961 674 SNHG7 - FANCD2 1220 694 1862 SNHG17 - FANCD2 1524 454 1198 SNHG16 - FANCD2 1600 2275 882 SNHG15 - FANCD2 1669 1464 322 SNHG1 - FANCD2 1705 2006 1185 SNHG18 - FANCD2 2276 614 1734 SNHG8 - FANCD2 2407 1232 2160 Table 39: 2nd order interaction ranking between FANCD2 VS SNHG family members. One can also interpret the results of the table 39 graphically, with the following influences - •SNHG family w.r.t FANCD2 with FANCD2 −>SNHG-3/6/10/5/7/17/15. UNEXPLORED COMBINATORIAL HYPOTHESES SNHG members w.r.t FANCD2 SNHG-3/6/10/5/7/17/15 FANCD2 Table 40: 2nd order combinatorial hypotheses between FANCD2 and SNHG family members. 23 3. Conclusion Presented here are a range of multiple synergistic FANCD2 2nd order combinations that were ranked via a machine learning based search engine. Via majority voting across the ranking methods, it was possible to find plausible unexplored synergistic combinations of FANCD2-X that might be prevalent in CRC cells after treatment with ETC-1922159 drug. Conflict of interest There are no conflicts to declare. Author’s contributions Concept, design, in silico implementation - SS. Analysis and interpretation of results - SS. Manuscript writing - SS. Manuscript revision - SS. Approval of manuscript - SS Acknowledgements Special thanks to Mrs. Rita Sinha and Mr. Prabhat Sinha for supporting the author financially, without which this work could not have been made possible. Source of Data Data used in this research work was released in a publication in Madan et al. [4]. 4. References References [1] A. D. D’Andrea, The fanconi road to cancer, Genes & development 17 (2003) 1933–1936. [2] S. Houghtaling, C. Timmers, M. Noll, M. J. Finegold, S. N. Jones, M. S. Meyn, M. Grompe, Epithelial cancer in fanconi anemia complementation group d2 (fancd2) knockout mice, Genes & development 17 (2003) 2021–2035. [3] S. van Twest, V. J. Murphy, C. Hodson, W. Tan, P. Swuec, J. J. ORourke, J. Heierhorst, W. Crismani, A. J. Deans, Mechanism of ubiquitination and deubiquitination in the fanconi anemia pathway, Molecular cell 65 (2017) 247– 259. [4] B. Madan, Z. Ke, N. Harmston, S. Y. Ho, A. Frois, J. Alam, D. A. Jeyaraj, V. Pendharkar, K. Ghosh, I. H. Virshup, et al., Wnt addiction of genetically defined cancers reversed by porcn inhibition, Oncogene 35 (2016) 2197. [5] S. Sinha, Machine learning ranking of plausible (un) explored synergistic gene combinations using sensitivity indices of time series measurements of wnt signaling pathway, Integrative Biology 16 (2024) zyae020. [6] S. Sinha, Sensitivity analysis based ranking reveals unknown biological hypotheses for down regulated genes in time buffer during administration of porcn-wnt inhibitor etc-1922159 in crc, bioRxiv (2017) 180927. 24 [7] T. Joachims, Training linear svms in linear time, in: Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining, ACM, 2006, pp. 217–226. [8] A. Bhat, P. L. Andersen, Z. Qin, W. Xiao, Rev3, the catalytic subunit of polζ, is required for maintaining fragile site stability in human cells, Nucleic acids research 41 (2013) 2328–2339. [9] P. Pichierri, A. Franchitto, F. Rosselli, Blm and the fanc proteins collaborate in a common pathway in response to stalled replication forks, The EMBO journal 23 (2004) 3154–3163. [10] C. Roques, Y. Coulombe, M. Delannoy, J. Vignard, S. Grossi, I. Brodeur, A. Rodrigue, J. Gautier, A. Z. Stasiak, A. Stasiak, et al., Mre11–rad50–nbs1 is a critical regulator of fancd2 stability and function during dna double-strand break repair, The EMBO journal 28 (2009) 2400–2413. [11] A. Kaur, J. Y. S. Lim, S. Sepramaniam, S. Patnaik, N. Harmston, M. A. Lee, E. Petretto, D. M. Virshup, B. Madan, Wnt inhibition creates a brca-like state in wnt-addicted cancer, EMBO molecular medicine 13 (2021) e13349. [12] C.-C. Liang, B. Zhan, Y. Yoshikawa, W. Haas, S. P. Gygi, M. A. Cohn, Uhrf1 is a sensor for dna interstrand crosslinks and recruits fancd2 to initiate the fanconi anemia pathway, Cell reports 10 (2015) 1947–1956. [13] C. S. Tremblay, F. F. Huang, O. Habi, C. C. Huard, C. Godin, G. L´ evesque, M. Carreau, Hes1 is a novel interactor of the fanconi anemia core complex, Blood, The Journal of the American Society of Hematology 112 (2008) 2062– 2070. [14] T. Taniguchi, I. Garcia-Higuera, P. R. Andreassen, R. C. Gregory, M. Grompe, A. D. D’Andrea, S-phase–specific interaction of the fanconi anemia protein, fancd2, with brca1 and rad51, Blood, The Journal of the American Society of Hematology 100 (2002) 2414–2420. [15] I. Garcia-Higuera, T. Taniguchi, S. Ganesan, M. S. Meyn, C. Timmers, J. Hejna, M. Grompe, A. D D’Andrea, Interaction of the fanconi anemia proteins and brca1 in a common pathway, Molecular cell 7 (2001) 249–262. [16] R. Roy, J. Chun, S. N. Powell, Brca1 and brca2: different roles in a common pathway of genome protection, Nature Reviews Cancer 12 (2012) 68–78. [17] K. M. McCabe, A. Hemphill, Y. Akkari, P. M. Jakobs, D. Pauw, S. B. Olson, R. E. Moses, M. Grompe, Ercc1 is required for fancd2 focus formation, Molecular genetics and metabolism 95 (2008) 66–73. [18] R. Kikuno, T. Nagase, M. Nakayama, H. Koga, N. Okazaki, D. Nakajima, O. Ohara, Huge: a database for human kiaa proteins, a 2004 update integrating hugeppi and rouge, Nucleic acids research 32 (2004) D502–D504. [19] C. MacKay, A.-C. D´ eclais, C. Lundin, A. Agostinho, A. J. Deans, T. J. MacArtney, K. Hofmann, A. Gartner, S. C. West, T. Helleday, et al., Identification of kiaa1018/fan1, a dna repair nuclease recruited to dna damage by monoubiquitinated fancd2, Cell 142 (2010) 65–76. [20] P. R. Andreassen, H. Hanenberg, Xrcc2 (x-ray repair cross complementing 2), Atlas of genetics and cytogenetics in oncology and haematology 23 (2019) 1. [21] F. Rossi, A. Helbling-Leclerc, R. Kawasumi, N. K. Jegadesan, X. Xu, P. Devulder, T. Abe, M. Takata, D. Xu, F. Rosselli, et al., Smc 5/6 acts jointly with fanconi anemia factors to support dna repair and genome stability, EMBO reports 21 (2020) e48222. [22] A. M. Gurtan, P. Stuckert, A. D. D’Andrea, The wd40 repeats of fancl are required for fanconi anemia core complex assembly, Journal of Biological Chemistry 281 (2006) 10896–10905. [23] Y. J. Machida, Y. Machida, Y. Chen, A. M. Gurtan, G. M. Kupfer, A. D. D’Andrea, A. Dutta, Ube2t is the e2 in the fanconi anemia pathway and undergoes negative autoregulation, Molecular cell 23 (2006) 589–596. [24] N. Zhang, R. Kaur, S. Akhter, R. J. Legerski, Cdc5l interacts with atr and is required for the s-phase cell-cycle checkpoint, EMBO reports 10 (2009) 1029–1035. [25] J. Wang, T. R. Sarkar, M. Zhou, S. Sharan, D. A. Ritt, T. D. Veenstra, D. K. Morrison, A.-M. Huang, E. Sterneck, Ccaat/enhancer binding protein delta (c/ebpδ, cebpd)-mediated nuclear import of fancd2 by ipo4 augments cellular response to dna damage, Proceedings of the National Academy of Sciences 107 (2010) 16131–16136. 25