Machine Learning Discoveries of FOXM1-X Synergy in ETC-1922159 Treated Colorectal Cancer Cells

Machine learning discoveries of FOXM1-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 Forkhead box protein M1 (FOXM1) belongs to the family of transcription factors Forkhead box (FOX). Forkhead box is a sequence of 80 to 100 amino acids forming a motif that binds to DNA. It is also known as the winged helix due to the butterflylike appearance of the loops in the protein structure of the domain. FOXM1 stimulates proliferation by promoting S-phase entry and M-phase entry and is also involved in proper execution of mitosis. Further, it is implicated in tumorigenesis and causes both tumor initiation and progression. In colorectal cancer (CRC) cells treated with ETC-1922159, FOXM1 was found to be down regulated along with other genes. A recently developed search engine ranked combinations of FOXM1-X (X, a particular gene/protein) at 2nd order level after drug administration. Some of these combinations have been tested in wet lab, however many have been pointed out by the search engine that are yet to be explored/tested. These rankings reveal which FOXM1X combinations might be working synergistically in CRC. In this research work, I cover combinations of FOXM1 with members of OTU deubiquitinase (OTUD), telomerase reverse transcriptase (TERT), RAD9-HUS1-RAD1 interacting nuclear orphan 1 (RHNO1), stathmin 1/oncoprotein 18 (STMN1), Wnt family member (WNT), TPX2 microtubule nucleation factor (TPX2), Never in mitosis gene A (NIMA)-related kinase 2 (NEK2), serine peptidase inhibitor Kazal type (SPINK), ATP binding cassette subfamily (ABC), alkB alkylation repair homolog (ALKBH), Rho GTPase activating proteins (ARHGAP), ASF1 anti-silencing function histone chaperone (ASF1), peroxiredoxin (PRDX), aurora kinase (AURK), cluster of differentiation/designation molecules (CD), cell division cycle (CDC), cell division cycle associated (CDCA), E2F transcription factor (E2F), cyclin dependent kinase (CDK), centromere protein (CENP), cytoskeleton associated protein (CKAP), DEAD-box helicase (DDX), maternal embryonic leucine zipper kinase (MELK), DEP domain containing (DEPDC), ERCC excision repair (ERCC), family with sequence similarity (FAM), Fanconi anemia complementation group (FANC), frizzled class receptor (FZD), high mobility group (HMG), IML dicoveries of FOXM1-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 29, 2025 heterogeneous nuclear ribonucleoprotein (HNRNP), heat shock protein family (HSP), kinesin family member (KIF), long intergenic non-protein coding RNA (LINC), methyltransferase (METTL), mitochondrial ribosomal protein L (MRPL), polo like kinase (PLK), DNA polymerase (POL), DNA-dependent/directed RNA polymerase (POLR), SCL/TAL1 interrupting locus (STIL), splicing factor 3 (SF3), ubiquitin specific peptidase (USP), integrin subunit (ITG), Zinc fingers C2H2-type family member (ZIC) and ubiquitin conjugating enzyme E2 (UBE2) family. Keywords: FOXM1, Porcupine inhibitor ETC-1922159, Sensitivity analysis, Machine learning, Colorectal cancer. 1. Introduction 1.1. FOXM1 Ye et al. [1] showed that hepatocyte nuclear factor 3α/β(HNF-3α/β) proteins had homology in the winged helix/fork head DNA binding domain and regulated cell-specific transcription in hepatocytes, and respiratory and intestinal epithelia. They described two isoforms of the winged helix transcription factor family, HNF-3/fork head homolog 11-A/B (HFH-11-A/B), isolated from the human colon carcinoma cell lines which were expressed in embryonic mesenchymal and epithelial cells and its expression was reactivated in adult cell types by proliferative signals or oxidative stress. Castaneda et al. [2] provide a review on the roles of many FOX family members in epithelial-mesenchymal transition (EMT), hormone signaling, drug resistance, metabolism, immune system regulation, cancer development and progression and their functions as pioneering factors. Wierstra and Alves [3] indicate that FOXM1 is a transcription factor that regulates genes controlling G1/S-transition, S-phase progression, G2/M-transition and M-phase progression. Invariably, both its expression and activity, are antagonistically regulated by many (anti- &) proliferation signals. Yang et al. [4] found that FOXM1 bound to and stimulated the promoter of Slug leading to EMT progression in human breast cancer. Francis et al. [5] observed that tyrosine kinase receptor HER2 receptor regulates the expression of the FOXM1 transcription factor, in breast cancer development. In prostate cancer cells, results by Liu et al. [6] provided new evidence for the regulatory mechanism of aberrant CDC6 oncogene transcription by FOXM1 and AR. In the study of pancreatic cancer metabolism, Cui et al. [7] found that FOXM1 regulates aerobic glycolysis (or Warburg effect, that is a shift from oxidative phosphorylation to glycolysis) by changing the expression of lactate dehydrogenase A (LDHA). This lead to elevated LDH activity, lactate production, and glucose utilization, thus causing cancer cell growth and metastasis. In gastric cancer cells, Zeng et al. [8] obseerved that FOXM1 was over-expressed and its inhibition prompted p53and p16independent senescence of cancer cells by regulating the expression of p27kip1and other targets. In colorectal cancer (CRC) cells treated with ETC-1922159, FOXM1 was found to be down regulated along with other genes. Some combinations of FOXM1 have been confirmed in wet lab, however, many of the combinations have not been explored/tested or 2 are known. To reveal these combinations, I use a modification of a recently published machine learning based search engine, details of which are given in the next section. 1.2. Combinatorial search problem and a possible solution In a recently published work Sinha [9], 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 [10]. The work uses SVM package by Joachims [11] in https://www.cs.cornell.edu/people/tj/svm_light/ svm_rank.html. I use the adaptation to rank 2nd order gene combinations. 2. Results & Discussion 2.1. FOXM1 related synergies 2.1.1. FOXM1 - OTUD6B / TERT / RHNO1 / STMN1 / WNT10B / TPX2 / NEK2 / SPINK4 In breast cancer cells Wang et al. [12] revealed that OTUD7B interacted with and deubiquitinated FOXM1, leading to its stabilization. This stabilization led to proliferation. They also showed that knocking down OTUD7B inhibited the proliferation and stemness of breast cancer cells by enhancing FOXM1 degradation. In gastric cancer, Tang et al. [13] found that FOXM1 interrupted the interaction between the E3 ligase MKRN1 and hTERT and decreased the protein degradation of the latter. Further studies revealed that FOXM1 interacted with hTERT through its DNA-binding domain (DBD) region. hTERT is the core subunit of telomerase that facilitates cancer initiation and progression by maintaining cell immortalization, promoting cell proliferation and inhibiting cell apoptosis. They found that hTERT played important roles in FOXM1-mediated activation of the Wnt/β-catenin pathway to promote gastric cancer cell proliferation. Thus they found a novel non-classical function of FOXM1 to increase hTERT protein stability. In high-grade serous carcinoma (HGSC), Barger et al. [14] showed that FOXM1 and RHNO1 are bidirectional genes (BDG) that are co-regulated by a bidirectional promoter (BDP) (named F/R-BDP). FOXM1 and RHNO1 each promoted oncogenic phenotypes in HGSC cells, that included clonogenic growth, DNA homologous recombination repair, and poly-ADP ribosylase inhibitor resistance. STMN1 is a microtubule-binding protein which inhibits the assembly of microtubule dimer or promotes the depolymerization of microtubules. It was reported as a major responsive factor of paclitaxel resistance for clinical chemotherapy of tumor patients. Liu et al. [15] suggest that FOXM1 promotes cell proliferation by upregulating STMN1 axis and contributes to tumorigenesis in gastric cancer, hepatocellular carcinoma and colorectal cancer. In cultured renal tubular epithelial cells, Xie et al. [16] found that overexpression of FOXM1 promoted 8 Wnts expression, while knock-down on FOXM1suppressed multi-Wnts including WNT1, WNT2b and WNT3 expression induced by 3 Ang II. Their findings show that FOXM1 regulated multi-Wnt family members leading to renal fibrosis. In hepatocellular carcinoma (HCC), Wang et al. [17] showed that FOXM1 bound to the TPX2 promoter, and regulated TPX2 activity to drive proliferation. In esophageal squamous cell carcinoma (ESCC)tissues, Li et al. [18] found that NEK2 was upregulated and FOXM1 was identified as its downstream target in the regulation of ESCC. Their mechanistic studies indicated that NEK2 may promote the malignant progression of ESCC by inhibiting cellular senescence through the activation of the FOXM1/c-Myc/p27 signaling pathways. Ding et al. [19] found that FOXM1 directly promotes SPINK1 transcription, enhancing tumor cell proliferation and metastasis in liver cancer, while regulating the p53 pathway. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these inidividual members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these individual members along with FOXM1. 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 these individual members w.r.t FOXM1. TERT - FOXM1 shows low ranking of 385 (laplace), 318 (linear) and 381 (rbf). RHNO1 - FOXM1 shows low ranking of 731 (linear) and 1564 (rbf). STMN1 - FOXM1 shows low ranking of 168 (laplace), 284 (linear) and 221 (rbf). WNT10B - FOXM1 shows low ranking of 1020 (laplace), 1313 (linear) and 560 (rbf). TPX2 - FOXM1 shows low ranking of 771 (laplace), 45 (linear) and 286 (rbf). NEK2 - FOXM1 shows low ranking of 492 (laplace), 144 (linear) and 350 (rbf). SPINK4 - FOXM1 shows low ranking of 972 (laplace) and 1042 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, OTUD6B showed high ranking and might not be synergistically working with FOXM1, before treatment. One can also interpret the results of the table 1 graphically, with the following influences - •individual members w.r.t FOXM1 with FOXM1 −>TERT / RHNO1 / STMN1 / WNT10B / TPX2 / NEK2 / SPINK4. 2.1.2. FOXM1 - ABC Paclitaxel is clinically used as a chemotherapeutic adminstration for several cancer types, but acquired drug resistance results in the failure of therapy, metastasis and relapse. The drug efflux mediated by ABC transporters and the survival signals activated by FOX molecules are influential in the development of paclitaxel drug resistance. Hou et al. [20] developed several types of paclitaxel-resistant (TR) nasopharyngeal carcinoma (NPC) cells which acquired cancer stem cell (CSC) phenotypes and underwent EMT, and developed multidrug resistance. They found that FOXM1 and ABCC5 were overexpressed in the TR NPC cells and patient tumor tissues as FOXM1 regulated ABCC5 gene transcription by binding to the FHK consensus motifs at the promoter region. Similar results were reported in cervical cancer cells by Hou et al. [21] and He et al. [22]. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC4 RANKING INDIVIDUAL MEMBERS VS FOXM1 RANKING OF INDIVIDUAL MEMBERS W.R.TFOXM1 laplace linear rbf OTUD6B - FOXM1 1706 1660 1389 TERT - FOXM1 385 318 381 RHNO1 - FOXM1 1874 731 1564 STMN1 - FOXM1 168 284 221 WNT10B - FOXM1 1020 1313 560 TPX2 - FOXM1 771 45 286 NEK2 - FOXM1 492 144 350 SPINK4 - FOXM1 972 2225 1042 Table 1: 2nd order interaction ranking between FOXM1 VS Individual members. UNEXPLORED COMBINATORIAL HYPOTHESES Individual members w.r.t FOXM1 TERT FOXM1 RHNO1 FOXM1 STMN1 FOXM1 WNT10B FOXM1 TPX2 FOXM1 NEK2 FOXM1 SPINK4 FOXM1 Table 2: 2nd order combinatorial hypotheses between FOXM1 and individual members. 1922159, these ABC members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these ABC members along with FOXM1. 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 ABC members w.r.t FOXM1.ABCE1 - FOXM1 shows low ranking of 464 (laplace), 493 (linear) and 885 (rbf). ABCA2 - FOXM1 shows 5 low ranking of 1019 (laplace), 1079 (linear) and 946 (rbf). ABCF2 - FOXM1 shows low ranking of 1531 (laplace) and 1432 (linear). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. RANKING ABC MEMBERS VS FOXM1 RANKING OF ABC MEMBERS W.R.TFOXM1 laplace linear rbf ABCE1 - FOXM1 464 493 885 ABCA2 - FOXM1 1019 1079 946 ABCF2 - FOXM1 1531 1432 2056 Table 3: 2nd order interaction ranking between FOXM1 VS ABC members. One can also interpret the results of the table 3 graphically, with the following influences - •ABC members w.r.t FOXM1 with FOXM1 −>ABC-E1/A2/F2. UNEXPLORED COMBINATORIAL HYPOTHESES ABC members w.r.t FOXM1 ABC-E1/A2/F2 FOXM1 Table 4: 2nd order combinatorial hypotheses between FOXM1 and ABC members. 2.1.3. FOXM1 - ALKBH In ovarian cancer (OV) cells, results by Li et al. [23] showed that ALKBH5 directly regulated the m6A modification and stability of PVT1 which further led to regulation of FOXM1, thus affecting malignant behaviours and chemosensitivity. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these ALKBH members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these ALKBH members along with FOXM1. 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 ALKBH members w.r.t FOXM1. ALKBH2 - FOXM1 shows low ranking of 1081 (laplace), 803 (linear) and 561 (rbf). ALKBH8 - FOXM1 shows low ranking of 770 (linear) and 1347 (rbf). These rankings point to the synergy 6 existing between the two components, which have been down regulated after the drug treatment. Further, ALKBH4 showed high ranking and might not be synergistically working with FOXM1, before treatment. RANKING ALKBH MEMBERS VS FOXM1 RANKING OF ALKBH MEMBERS W.R.TFOXM1 laplace linear rbf ALKBH2 - FOXM1 1081 803 561 ALKBH8 - FOXM1 1648 770 1347 ALKBH4 - FOXM1 2285 2575 2568 Table 5: 2nd order interaction ranking between FOXM1 VS ALKBH members. One can also interpret the results of the table 5 graphically, with the following influences - •ALKBH members w.r.t FOXM1 with FOXM1 −>ALKBH-2/8. UNEXPLORED COMBINATORIAL HYPOTHESES ALKBH members w.r.t FOXM1 ALKBH-2/8 FOXM1 Table 6: 2nd order combinatorial hypotheses between FOXM1 and ALKBH members. 2.1.4. FOXM1 - ARHGAP ARHGAP11A can facilitate GTP hydrolysis in RhoA and is transcriptionally activated by FOXM1 via multiple FOXM1 binding sites in the promoter regions in tongue squamous cell carcinoma (TSCC) cells, as shown by Zhang et al. [24]. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these ARHGAP members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these ARHGAP members along with FOXM1. 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 ARHGAP members w.r.t FOXM1. ARHGAP11A - FOXM1 shows low ranking of 36 (laplace), 123 (linear) and 12 (rbf). ARHGAP11B - FOXM1 shows low ranking of 499 (laplace), 36 (linear) and 54 (rbf). ARHGAP33 - FOXM1 shows low ranking of 1013 (laplace), 1036 (linear) and 354 (rbf). These rankings point 7 to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, ARHGAP19 showed high ranking and might not be synergistically working with FOXM1, before treatment. RANKING ARHGAP MEMBERS VS FOXM1 RANKING OF ARHGAP MEMBERS W.R.TFOXM1 laplace linear rbf ARHGAP11A - FOXM1 36 123 12 ARHGAP11B - FOXM1 499 36 54 ARHGAP33 - FOXM1 1013 1036 354 ARHGAP19 - FOXM1 2519 1627 1846 Table 7: 2nd order interaction ranking between FOXM1 VS ARHGAP members. One can also interpret the results of the table 7 graphically, with the following influences - •ARHGAP members w.r.t FOXM1 with FOXM1 −>ARHGAP-11A/11B/33. UNEXPLORED COMBINATORIAL HYPOTHESES ARHGAP members w.r.t FOXM1 ARHGAP-11A/11B/33 FOXM1 Table 8: 2nd order combinatorial hypotheses between FOXM1 and ARHGAP members. 2.1.5. FOXM1 - ASF/PRDX In gastric cancer (GC), Zhao et al. [25] showed that FOXM1 transcriptionally activates ASF1B and the latter is a downstream effector of FOXM1-mediated progression. Further this proliferation and maintenance of oxidative stress homeostasis in GC happens via ASF1B-induced downstream PRDX3 expression. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these ASF1-PRDX members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these ASF1-PRDX members along with FOXM1. 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 ASF-PRDX members w.r.t FOXM1. ASF1B - FOXM1 8 shows low ranking of 569 (laplace), 404 (linear) and 663 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, ASF1A, PRDX3, PRDX4, PRDX2 and PRDX6 showed high ranking and might not be synergistically working with FOXM1, before treatment. With respect to the above reference in gastric cancer, ASF1 and PRDX3 did not show synergistic role in CRC before treatment, as pointed out by the high rank in CRC treated with ETC-1922159. RANKING ASF1-PRDX MEMBERS VS FOXM1 RANKING OF ASF1-PRDX MEMBERS W.R.TFOXM1 laplace linear rbf ASF1B - FOXM1 569 404 663 ASF1A - FOXM1 1728 2294 1945 PRDX3 - FOXM1 1730 1943 1603 PRDX4 - FOXM1 1958 1339 2363 PRDX2 - FOXM1 2230 2505 1886 PRDX6 - FOXM1 2622 1790 2567 Table 9: 2nd order interaction ranking between FOXM1 VS ASF1-PRDX members. One can also interpret the results of the table 9 graphically, with the following influences - •ASF1-PRDX members w.r.t FOXM1 with FOXM1 −>ASF1-B. UNEXPLORED COMBINATORIAL HYPOTHESES ASF1-PRDX members w.r.t FOXM1 ASF1-B FOXM1 Table 10: 2nd order combinatorial hypotheses between FOXM1 and ASF1-PRDX members. 2.1.6. FOXM1 - AURK In a review, Chen et al. [26] indicate that AURKA can function in either the cytoplasm or the nucleus and regulates basic cellular processes. This regulation happens via the phosphorylation of downstream substrates and AURKA can promote the transcription and expression of oncogenes together with other transcription factors in the nucleus like FOXM1. These experimental studies confirm the existence of synergy between 9 UNEXPLORED COMBINATORIAL HYPOTHESES CDK members w.r.t FOXM1 CDK-1/20 FOXM1 Table 20: 2nd order combinatorial hypotheses between FOXM1 and CDK members. sues. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these CENP members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these CENP members along with FOXM1. Table 21 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 22 generated from analysis of the ranks in table 21. The table 21 shows rankings of CENP members w.r.t FOXM1. CENPH - FOXM1 shows low ranking of 49 (laplace), 223 (linear) and 48 (rbf). CENPU - FOXM1 shows low ranking of 225 (laplace), 145 (linear) and 80 (rbf). CENPA - FOXM1 shows low ranking of 245 (laplace), 113 (linear) and 174 (rbf). CENPE - FOXM1 shows low ranking of 257 (laplace), 88 (linear) and 59 (rbf). CENPF - FOXM1 shows low ranking of 342 (laplace), 227 (linear) and 57 (rbf). CENPJ - FOXM1 shows low ranking of 435 (laplace), 585 (linear) and 74 (rbf). CENPM - FOXM1 shows low ranking of 463 (laplace), 240 (linear) and 868 (rbf). CENPL - FOXM1 shows low ranking of 478 (laplace) and 948 (rbf). CENPN - FOXM1 shows low ranking of 511 (laplace), 856 (linear) and 1378 (rbf). CENPW - FOXM1 shows low ranking of 531 (laplace), 725 (linear) and 445 (rbf). CENPK - FOXM1 shows low ranking of 612 (laplace), 116 (linear) and 573 (rbf). CENPI - FOXM1 shows low ranking of 920 (laplace), 491 (linear) and 563 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, CENPO and CENPV showed high ranking and might not be synergistically working with FOXM1, before treatment. RANKING CENP MEMBERS VS FOXM1 RANKING OF CENP MEMBERS W.R.TFOXM1 laplace linear rbf laplace linear rbf CENPH - FOXM1 49 223 48 CENPU - FOXM1 225 145 80 CENPA - FOXM1 245 113 174 CENPE - FOXM1 257 88 59 CENPF - FOXM1 342 227 57 CENPJ - FOXM1 435 585 74 CENPM - FOXM1 463 240 868 CENPL - FOXM1 478 1758 948 CENPN - FOXM1 511 856 1378 CENPW - FOXM1 531 725 445 CENPK - FOXM1 612 116 573 CENPI - FOXM1 920 491 563 CENPO - FOXM1 1968 1694 2003 CENPV - FOXM1 2293 2638 1030 Table 21: 2nd order interaction ranking between FOXM1 VS CENP members. 16 One can also interpret the results of the table 21 graphically, with the following influences - •CENP members w.r.t FOXM1 with FOXM1 −>CENP-H/U/A/E/F/J/M/L/N/W/K/I. UNEXPLORED COMBINATORIAL HYPOTHESES CENP members w.r.t FOXM1 CENP-H/U/A/E/F/J/M/L/N/W/K/I FOXM1 Table 22: 2nd order combinatorial hypotheses between FOXM1 and CENP members. 2.1.12. FOXM1 - CKAP CKAP4 has been reported as an important regulator of glioblastoma (GBM). Xu et al. [32] demostrated that CKAP4 knockdown remarkably reduced the malignant potential of GBM cells, whereas its overexpression had the reverse effects. Further they showed that overexpression of CKAP4 led to increased FOXM1 expression in union with an increased level of AKT and ERK phosphorylation. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these CKAP members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these CKAP members along with FOXM1. Table 23 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 24 generated from analysis of the ranks in table 23. The table 23 shows rankings of CKAP members w.r.t FOXM1. CKAP2L - FOXM1 shows low ranking of 280 (laplace), 275 (linear) and 363 (rbf). CKAP2 - FOXM1 shows low ranking of 536 (laplace), 251 (linear) and 191 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, CKAP5 showed high ranking and might not be synergistically working with FOXM1, before treatment. RANKING CKAP MEMBERS VS FOXM1 RANKING OF CKAP MEMBERS W.R.TFOXM1 laplace linear rbf CKAP2L - FOXM1 280 275 363 CKAP2 - FOXM1 536 251 191 CKAP5 - FOXM1 1836 2446 2633 Table 23: 2nd order interaction ranking between FOXM1 VS CKAP members. 17 One can also interpret the results of the table 23 graphically, with the following influences - •CKAP members w.r.t FOXM1 with FOXM1 −>CKAP-2/2L. UNEXPLORED COMBINATORIAL HYPOTHESES CKAP members w.r.t FOXM1 CKAP-2/2L FOXM1 Table 24: 2nd order combinatorial hypotheses between FOXM1 and CKAP members. 2.1.13. FOXM1 - DDX In ovarian cancer (OC) cells, Zhao et al. [33] determined that DDX23 was upregulatedand by functional assays showed that DDX23 silencing impeded cell proliferation/invasion in vitro and tumor growth in vivo. They observed that DDX23 regulated the mRNA processing of FOXM1 and its silencing reduced the production of FOXM1C, an oncogenic transcript of FOXM1 in OC, thereby decreasing the FOXM1 protein expression and attenuating the malignant progression of OC. Similarly, in hepatocellular carcinoma (HCC) Li et al. [34] found DDX56 was found to be overexpressed, lead to tumor cell proliferation, migration, invasion, EMT, and stemness. They also showed that MELK mediated FOXM1 expression, thus regulating cancer stemness and malignant traits. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC1922159, these DDX members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these DDX members along with FOXM1. 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 DDX members w.r.t FOXM1. DDX20 - FOXM1 shows low ranking of 154 (laplace), 1370 (linear) and 1419 (rbf). DDX12P - FOXM1 shows low ranking of 296 (laplace), 347 (linear) and 207 (rbf). DDX28 - FOXM1 shows low ranking of 665 (laplace), 677 (linear) and 1980 DDX10 - FOXM1 shows low ranking of 1048 (laplace), 494 (linear) and 377 (rbf). DDX31 - FOXM1 shows low ranking of 1086 (laplace), 1463 (linear) and 971 (rbf). DDX21 - FOXM1 shows low ranking of 1272 (laplace) and 1116 (rbf). DDX55 - FOXM1 shows low ranking of 1287 (laplace), 915 (linear) and 1440 (rbf). DDX11 - FOXM1 shows low ranking of 1289 (laplace), 1775 and 875 (rbf). DDX51 - FOXM1 shows low ranking of 1307 (linear) and 1070 (rbf). MELK - FOXM1 shows low ranking of 961 (laplace), 461 (linear) and 237 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, DDX19A, DDX56, DDX46, DDX18, DDX27 and DDX54 showed high ranking and might not be synergistically working with FOXM1, before treatment. One can also interpret the results of the table 25 graphically, with the following influences - •DDX members w.r.t FOXM1 with FOXM1 −>DDX-20/12P/28/10/31/21/55/11/51 18 RANKING DDX MEMBERS VS FOXM1 RANKING OF DDX MEMBERS W.R.TFOXM1 laplace linear rbf laplace linear rbf DDX20 - FOXM1 154 1370 1419 DDX12P - FOXM1 296 347 207 DDX28 - FOXM1 665 677 1980 DDX10 - FOXM1 1048 494 377 DDX31 - FOXM1 1086 1463 971 DDX21 - FOXM1 1272 2005 1116 DDX55 - FOXM1 1287 915 1440 DDX11 - FOXM1 1289 1775 875 DDX51 - FOXM1 1629 1307 1070 DDX19A - FOXM1 1734 2126 1535 DDX56 - FOXM1 2363 2203 2269 DDX46 - FOXM1 2471 2149 2368 DDX18 - FOXM1 2523 1929 1585 DDX27 - FOXM1 2565 2055 2737 DDX54 - FOXM1 2608 2144 2528 MELK - FOXM1 961 461 237 Table 25: 2nd order interaction ranking between FOXM1 VS DDX members. and •MELK members w.r.t FOXM1 with FOXM1 −>MELK. UNEXPLORED COMBINATORIAL HYPOTHESES DDX members w.r.t FOXM1 DDX-20/12P/28/10/31/21/55/11/51 FOXM1 MELK FOXM1 Table 26: 2nd order combinatorial hypotheses between FOXM1 and DDX members. 2.1.14. FOXM1 - DEPDC Qiu et al. [35] identified elevated DEPDC1 expression in oral squamous cell carcinoma (OSCC). Additionally, FOXM1 interacted with DEPDC1, thus facilitating Wnt/β-catenin signal transduction and β-catenin protein nuclear expression. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these DEPDC members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these DEPDC members along with FOXM1. 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 DEPDC members w.r.t FOXM1. DEPDC4 - FOXM1 shows low ranking of 292 (laplace), 445 (linear) and 761 (rbf). DEPDC1B - FOXM1 shows low ranking of 397 (laplace), 224 (linear) and 106 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. 19 Further, DEPDC7 showed high ranking and might not be synergistically working with FOXM1, before treatment. RANKING DEPDC MEMBERS VS FOXM1 RANKING OF DEPDC MEMBERS W.R.TFOXM1 laplace linear rbf DEPDC4 - FOXM1 292 445 761 DEPDC1B - FOXM1 397 224 106 DEPDC7 - FOXM1 2138 1936 1920 Table 27: 2nd order interaction ranking between FOXM1 VS DEPDC members. One can also interpret the results of the table 27 graphically, with the following influences - •DEPDC members w.r.t FOXM1 with FOXM1 −>DEPDC-4/1B. UNEXPLORED COMBINATORIAL HYPOTHESES DEPDC members w.r.t FOXM1 DEPDC-4/1B FOXM1 Table 28: 2nd order combinatorial hypotheses between FOXM1 and DEPDC members. 2.1.15. FOXM1 - ERCC In laryngeal squamous cell carcinoma (LSCC), Cui et al. [36] revealed that ERCC6L expression was elevated and ERCC6L knockdown LSCC cells showed decreased proliferation and migration, increased apoptosis, and reactive oxygen species (ROS). Further, they showed that overexpression of ERCC6L caused nuclear translocation of FOXM1 to facilitate direct binding to the KIF4A promoter and upregulated KIF4A expression which lead to progression. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these ERCC members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these ERCC members along with FOXM1. 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 ERCC members w.r.t FOXM1. ERCC6L - FOXM1 shows low ranking of 153 (laplace), 82 (linear) and 193 (rbf). ERCC8 - FOXM1 shows low ranking of 870 (laplace), 1189 (linear) and 912 (rbf). These rankings point to the 20 synergy existing between the two components, which have been down regulated after the drug treatment. RANKING ERCC MEMBERS VS FOXM1 RANKING OF ERCC MEMBERS W.R.TFOXM1 laplace linear rbf ERCC6L - FOXM1 153 82 193 ERCC8 - FOXM1 870 1189 912 Table 29: 2nd order interaction ranking between FOXM1 VS ERCC members. One can also interpret the results of the table 29 graphically, with the following influences - •ERCC members w.r.t FOXM1 with FOXM1 −>ERCC-6L/8. UNEXPLORED COMBINATORIAL HYPOTHESES ERCC members w.r.t FOXM1 ERCC-6L/8 FOXM1 Table 30: 2nd order combinatorial hypotheses between FOXM1 and ERCC members. 2.1.16. FOXM1 - FAM Zhao et al. [37] reported higher expression levels of FAM64A in head and neck squamous cell carcinoma (HNSCC) tissues and cell lines. They found the existence of a physical interaction between FAM64A and FOXM1. FAM64A promoted tumorigenesis by enhancing the transcriptional activity of FOXM1, as well as by modulating FOXM1 expression via the autoregulation loop. Fu et al. [38] found that FAM72A is a cell-cycle-regulated gene that is transcriptionally and post-transcriptionally regulated by FOXM1 and APC/C, respectively. Study by Fei et al. [39], revealed that FAM83A hijacked the promoter of FOXM1 to progress the malignant lung adenocarcinoma (LUAD), and uncovered that the function of FAM83A is partly dependent on FOXM1 regulation. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these FAM members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these FAM members along with FOXM1. 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 FAM members w.r.t FOXM1. FAM83D - FOXM1 21 shows low ranking of 3 (laplace), 130 (linear) and 515 (rbf). FAM111B - FOXM1 shows low ranking of 39 (laplace), 143 (linear) and 93 (rbf). FAM72D - FOXM1 shows low ranking of 79 (laplace), 208 (linear) and 139 (rbf). FAM201A - FOXM1 shows low ranking of 304 (laplace), 389 (linear) and 729 (rbf). FAM169A - FOXM1 shows low ranking of 312 (laplace), 267 (linear) and 210 (rbf). FAM96AP2 - FOXM1 shows low ranking of 320 (laplace), 875 (linear) and 658 (rbf). FAM216A - FOXM1 shows low ranking of 348 (laplace), 538 (linear) and 617 (rbf). FAM120C - FOXM1 shows low ranking of 361 (laplace), 218 (linear) and 413 (rbf). FAM161A - FOXM1 shows low ranking of 564 (laplace), 9 (linear) and 243 (rbf). FAM227A - FOXM1 shows low ranking of 568 (laplace), 490 (linear) and 351 (rbf). FAM72A - FOXM1 shows low ranking of 673 (laplace), 586 (linear) and 597 (rbf). FAM131B - FOXM1 shows low ranking of 832 (laplace), 956 (linear) and 664 (rbf). FAM81A - FOXM1 shows low ranking of 918 (laplace), 155 (linear) and 403 (rbf). FAM210B - FOXM1 shows low ranking of 1148 (laplace), 457 (linear) and 1153 (rbf). FAM96A - FOXM1 shows low ranking of 1449 (laplace), 1158 (linear) and 1213 (rbf). FAM98B - FOXM1 shows low ranking of 1474 (laplace), 812 (linear) and 670 (rbf). FAM86JP - FOXM1 shows low ranking of 1408 (linear) and 1297 (rbf). FAM168A - FOXM1 shows low ranking of 1182 (linear) and 835 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, FAM86A, FAM89A, FAM86C2P, FAM185A, FAM221A, FAM122B, FAM117A, FAM86B1, FAM208B, FAM173B, FAM117B, FAM86C1, FAM98A, FAM136A, FAM149B1 and FAM178A showed high ranking and might not be synergistically working with FOXM1, before treatment. RANKING FAM MEMBERS VS FOXM1 RANKING OF FAM MEMBERS W.R.TFOXM1 laplace linear rbf laplace linear rbf FAM83D - FOXM1 3 130 515 FAM111B - FOXM1 39 143 93 FAM72D - FOXM1 79 208 139 FAM201A - FOXM1 304 389 729 FAM169A - FOXM1 312 267 210 FAM96AP2 - FOXM1 320 875 658 FAM216A - FOXM1 348 538 617 FAM120C - FOXM1 361 218 413 FAM161A - FOXM1 564 9 243 FAM227A - FOXM1 568 490 351 FAM72A - FOXM1 673 586 597 FAM131B - FOXM1 832 956 664 FAM81A - FOXM1 918 155 403 FAM86A - FOXM1 967 2469 1994 FAM210B - FOXM1 1148 457 1153 FAM96A - FOXM1 1449 1158 1213 FAM98B - FOXM1 1474 812 670 FAM89A - FOXM1 1566 2115 1226 FAM86C2P - FOXM1 1615 1819 2159 FAM185A - FOXM1 1762 1617 2098 FAM221A - FOXM1 1777 1050 1634 FAM86JP - FOXM1 1881 1408 1297 FAM122B - FOXM1 1921 2502 2134 FAM117A - FOXM1 1927 1474 2423 FAM168A - FOXM1 1931 1182 835 FAM86B1 - FOXM1 2025 1108 2158 FAM208B - FOXM1 2109 1701 1841 FAM173B - FOXM1 2331 1680 2561 FAM117B - FOXM1 2332 2445 2041 FAM86C1 - FOXM1 2336 2469 2098 FAM98A - FOXM1 2534 2477 2467 FAM136A - FOXM1 2603 2691 2325 FAM149B1 - FOXM1 2668 2612 2602 FAM178A - FOXM1 2739 2302 2693 Table 31: 2nd order interaction ranking between FOXM1 VS FAM members. One can also interpret the results of the table 31 graphically, with the following 22 influences - •FAM members w.r.t FOXM1 with FOXM1 −>FAM-83D / 111B / 72D / 201A / 169A / 96AP2 / 216A / 120C / 161A / 227A / 72A / 131B / 81A / 210B / 96A / 98B / 86JP / 168A. UNEXPLORED COMBINATORIAL HYPOTHESES FAM members w.r.t FOXM1 FAM-83D/111B/72D/201A/169A/96AP2/216A/120C/... 161A/227A/72A/131B/81A/210B/96A/98B/86JP/168A FOXM1 Table 32: 2nd order combinatorial hypotheses between FOXM1 and FAM members. 2.1.17. FOXM1 - FANC In nonmuscle invasive bladder cancer (NMIBC), Roh et al. [40] found that FOXM1 and FANCD2 were involved in recurrence. Their investigation showed that FOXM1 directly regulated the transcription of FANCD2 and depletion of FOXM1 resulted in DNA repair defects in the FA pathway and decreased resistance to chemotherapy. Thus, the FANCD2-associated FA pathway activation by FOXM1 is critial to drug resistance and bladder cancer recurrence. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these FANC members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these FANC members along with FOXM1. 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 FANC members w.r.t FOXM1. FANCB - FOXM1 shows low ranking of 90 (laplace), 29 (linear) and 196 (rbf). FANCM - FOXM1 shows low ranking of 114 (laplace), 950 (linear) and 516 (rbf). FANCG - FOXM1 shows low ranking of 591 (laplace), 690 (linear) and 1136 (rbf). FANCI - FOXM1 shows low ranking of 1118 (laplace) and 1467 (linear) FANCD2 - FOXM1 shows low ranking of 1446 (laplace), 967 (linear) and 396 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, FANCF showed high ranking and might not be synergistically working with FOXM1, before treatment. One can also interpret the results of the table 33 graphically, with the following influences - •FANC members w.r.t FOXM1 with FOXM1 −>FANC-B/M/G/I/D2. 2.1.18. FOXM1 - FZD Triple-negative breast cancer (TNBC) is known to develop chemoresistance, which is responsible for cancer recurrence and distal metastasis. Both DNA damage repair and stemness are related to chemoresistance. Sun et al. [41] identified FZD5 as expressed in TNBC and revealed that it contributed to TNBC cell G1/S transition, DNA replication, DNA damage repair, survival, and stemness. They show that FOXM1, which promoted 23 RANKING FANC MEMBERS VS FOXM1 RANKING OF FANC MEMBERS W.R.TFOXM1 laplace linear rbf FANCB - FOXM1 90 29 196 FANCM - FOXM1 114 950 516 FANCG - FOXM1 591 690 1136 FANCI - FOXM1 1118 1467 2293 FANCD2 - FOXM1 1446 967 396 FANCF - FOXM1 2484 2518 1900 Table 33: 2nd order interaction ranking between FOXM1 VS FANC members. UNEXPLORED COMBINATORIAL HYPOTHESES FANC members w.r.t FOXM1 FANC-B/M/G/I/D2 FOXM1 Table 34: 2nd order combinatorial hypotheses between FOXM1 and FANC members. BRCA1 and BIRC5 transcription, acted as a downstream effecter of FZD5 signaling. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these FZD members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these FZD members along with FOXM1. 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. The table 35 shows rankings of FZD members w.r.t FOXM1. FZD3 - FOXM1 shows low ranking of 590 (laplace), 940 (linear) and 1481 (rbf). FZD7 - FOXM1 shows low ranking of 974 (laplace), 470 (linear) and 1215 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. One can also interpret the results of the table 35 graphically, with the following influences - •FZD members w.r.t FOXM1 with FOXM1 −>FZD-3/7. 24 RANKING FZD MEMBERS VS FOXM1 RANKING OF FZD MEMBERS W.R.TFOXM1 laplace linear rbf FZD3 - FOXM1 590 940 1481 FZD7 - FOXM1 974 470 1215 Table 35: 2nd order interaction ranking between FOXM1 VS FZD members. UNEXPLORED COMBINATORIAL HYPOTHESES FZD members w.r.t FOXM1 FZD-3/7 FOXM1 Table 36: 2nd order combinatorial hypotheses between FOXM1 and FZD members. 2.1.19. FOXM1 - HMG Zanin et al. [42] discovered FOXM1 as a molecular partner of HMGA1 in regulating a gene network implicated in breast cancer hallmarks. They observed that HMGA1 formed a complex with FOXM1 and stabilized it in the nucleus, thus increasing its transcriptional activity on common target genes. Further, they demonstrated that HMGA1 and FOXM1 synergistically drove breast cancer cells to promote tumor angiogenesis both in vitro in endothelial cells and in vivo in a zebrafish xenograft model. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these HMG members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these HMG members along with FOXM1. 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 HMG members w.r.t FOXM1. HMGN5 - FOXM1 shows low ranking of 37 (laplace), 381 (linear) and 170 (rbf). HMGN3 - FOXM1 shows low ranking of 148 (laplace), 89 (linear) and 87 (rbf). HMGB2 - FOXM1 shows low ranking of 481 (laplace), 415 (linear) and 282 (rbf). HMGN2 - FOXM1 shows low ranking of 957 (laplace), 1151 (linear) and 529 (rbf). HMGB3 - FOXM1 shows low ranking of 1224 (laplace), 1064 (linear) and 1149 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, HMGB1 showed high ranking and might not be synergistically working with FOXM1, before treatment. 25 The table 47 shows rankings of METTL members w.r.t FOXM1. METTL1 - FOXM1 shows low ranking of 356 (laplace), 790 (linear) and 160 (rbf). METTL13 - FOXM1 shows low ranking of 719 (laplace) and 827 (linear). METTL21B - FOXM1 shows low ranking of 1010 (laplace) and 1079 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, METTL21A, METTL12, METTL3, METTL8, METTL16, METTL5, METTL17 and METTL2B showed high ranking and might not be synergistically working with FOXM1, before treatment. RANKING METTL MEMBERS VS FOXM1 RANKING OF METTL MEMBERS W.R.TFOXM1 laplace linear rbf laplace linear rbf METTL1 - FOXM1 356 790 160 METTL13 - FOXM1 719 827 1989 METTL21B - FOXM1 1010 1990 1079 METTL21A - FOXM1 1434 2342 2542 METTL12 - FOXM1 1608 1516 2224 METTL3 - FOXM1 2060 2574 1887 METTL8 - FOXM1 2168 2150 1144 METTL16 - FOXM1 2381 1469 1791 METTL5 - FOXM1 2478 2060 2320 METTL17 - FOXM1 2546 2614 2405 METTL2B - FOXM1 2574 2509 2574 Table 47: 2nd order interaction ranking between FOXM1 VS METTL members. One can also interpret the results of the table 47 graphically, with the following influences - •METTL members w.r.t FOXM1 with FOXM1 −>METTL-1/13/21B. UNEXPLORED COMBINATORIAL HYPOTHESES METTL members w.r.t FOXM1 METTL-1/13/21B FOXM1 Table 48: 2nd order combinatorial hypotheses between FOXM1 and METTL members. 2.1.25. FOXM1 - MRPL In LUAD, Zhang et al. [49] showed that MRPL51 knockdown suppressed cell proliferation, induced G1 phase arrest and decreased cell invasion. Further, it was transcriptionally activated by FOXM1 in LUAD and contributed to the malignant behaviors of tumor cells, including EMT, cell cycle progression and invasion. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these MRPL members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these MRPL members along with FOXM1. Table 49 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 50 generated from analysis of the ranks in table 49. 32 The table 49 shows rankings of MRPL members w.r.t FOXM1. MRPL16 - FOXM1 shows low ranking of 331 (laplace), 1087 (linear) and 606 (rbf). MRPL4 - FOXM1 shows low ranking of 1090 (laplace) and 1439 (rbf). MRPL11 - FOXM1 shows low ranking of 1117 (laplace), 444 (linear) and 983 (rbf). MRPL32 - FOXM1 shows low ranking of 1274 (laplace) and 998 (rbf). MRPL36 - FOXM1 shows low ranking of 1297 (laplace), 1014 (linear) and 997 (rbf). MRPL35 - FOXM1 shows low ranking of 1317 (laplace) and 1014 (rbf). MRPL3 - FOXM1 shows low ranking of 236 (linear) and 806 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, MRPL37, MRPL1, MRPL45, MRPL18, MRPL13, MRPL2, MRPL34, MRPL21, MRPL17, MRPL24, MRPL9, MRPL19, MRPL48, MRPL42, MRPL51, MRPL40, MRPL50, MRPL47 and MRPL22 showed high ranking and might not be synergistically working with FOXM1, before treatment. RANKING MRPL MEMBERS VS FOXM1 RANKING OF MRPL MEMBERS W.R.TFOXM1 laplace linear rbf laplace linear rbf MRPL16 - FOXM1 331 1087 606 MRPL4 - FOXM1 1090 1718 1439 MRPL11 - FOXM1 1117 444 983 MRPL32 - FOXM1 1274 2630 998 MRPL36 - FOXM1 1297 1014 997 MRPL35 - FOXM1 1317 2243 1014 MRPL37 - FOXM1 1496 2455 2254 MRPL1 - FOXM1 1527 1778 2466 MRPL45 - FOXM1 1581 2658 2208 MRPL18 - FOXM1 1590 2395 2344 MRPL3 - FOXM1 1618 236 806 MRPL13 - FOXM1 1803 860 1935 MRPL2 - FOXM1 1811 1373 2652 MRPL34 - FOXM1 1825 2143 882 MRPL21 - FOXM1 1842 1956 1942 MRPL17 - FOXM1 1866 1430 2100 MRPL24 - FOXM1 1872 2377 819 MRPL9 - FOXM1 1924 1801 2685 MRPL19 - FOXM1 2226 1852 2017 MRPL48 - FOXM1 2248 2255 1992 MRPL42 - FOXM1 2414 885 2553 MRPL51 - FOXM1 2440 2465 1939 MRPL40 - FOXM1 2464 2326 2099 MRPL50 - FOXM1 2533 2095 1669 MRPL47 - FOXM1 2645 2653 2656 MRPL22 - FOXM1 2657 2522 2732 Table 49: 2nd order interaction ranking between FOXM1 VS MRPL members. One can also interpret the results of the table 49 graphically, with the following influences - •MRPL members w.r.t FOXM1 with FOXM1 −>MRPL-16/4/11/32/36/35/3. UNEXPLORED COMBINATORIAL HYPOTHESES MRPL members w.r.t FOXM1 MRPL-16/4/11/32/36/35/3 FOXM1 Table 50: 2nd order combinatorial hypotheses between FOXM1 and MRPL members. 33 2.1.26. FOXM1 - PLK In LUAD, Xu et al. [50] showed that PLK1 phosphorylates FOXM1 at Ser25, thus acquiring the reprogramming ability to stimulate the invasive traits in cancer and influence immune cell plasticity. This invasive form of p-FOXM1 upregulates the expression of IL1A/1B, VEGFA, and IL6 by direct activation, recruiting monocytes and promoting the polarization of M2d-like tumor-associated macrophages (TAMs). These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these PLK members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these PLK members along with FOXM1. Table 51 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 52 generated from analysis of the ranks in table 51. The table 51 shows rankings of PLK members w.r.t FOXM1. PLK4 - FOXM1 shows low ranking of 506 (laplace), 253 (linear) and 258 (rbf). PLK1 - FOXM1 shows low ranking of 1108 (laplace), 1082 (linear) and 1007 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. RANKING PLK MEMBERS VS FOXM1 RANKING OF PLK MEMBERS W.R.TFOXM1 laplace linear rbf PLK4 - FOXM1 506 253 258 PLK1 - FOXM1 1108 1082 1007 Table 51: 2nd order interaction ranking between FOXM1 VS PLK members. One can also interpret the results of the table 51 graphically, with the following influences - •PLK members w.r.t FOXM1 with FOXM1 −>PLK-4/1. UNEXPLORED COMBINATORIAL HYPOTHESES PLK members w.r.t FOXM1 PLK-4/1 FOXM1 Table 52: 2nd order combinatorial hypotheses between FOXM1 and PLK members. 34 2.1.27. FOXM1 - POL/POLR In LUAD, Ni et al. [51] showed that berberine, a natural anticancer drug, interfered the expression of survival related gene POLE2 in DNA replication mediated by FOXM1. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these POL members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these POL members along with FOXM1. Additionally, I also generated rankings of members of POLR to see if there was synergy between them and FOXM1. Table 53 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 54 generated from analysis of the ranks in table 53. The table 53 shows rankings of POL members w.r.t FOXM1. POLQ - FOXM1 shows low ranking of 16 (laplace), 11 (linear) and 136 (rbf). POLE2 - FOXM1 shows low ranking of 131 (laplace), 801 (linear) and 533 (rbf). POLA1 - FOXM1 shows low ranking of 593 (laplace), 771 (linear) and 82 (rbf). POLG2 - FOXM1 shows low ranking of 749 (laplace), 1396 (linear) and 831 (rbf). POLD1 - FOXM1 shows low ranking of 753 (laplace), 409 (linear) and 611 (rbf). POLD2 - FOXM1 shows low ranking of 1052 (laplace) and 435 (rbf). POLR3K - FOXM1 shows low ranking of 609 (laplace), 868 (linear) and 493 (rbf). POLR1C - FOXM1 shows low ranking of 845 (laplace), 850 (linear) and 1218 (rbf). POLR2G - FOXM1 shows low ranking of 1147 (laplace), 1074 (linear) and 1739 POLR1B - FOXM1 shows low ranking of 1401 (laplace), 705 (linear) and 673 (rbf). POLR1A - FOXM1 shows low ranking of 1022 (linear) and 1188 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, POLE3, POLA2, POLB, POLR1E, POLR3E, POLR2D, POLR2F, POLR1D, POLR3A and POLR2K showed high ranking and might not be synergistically working with FOXM1, before treatment. One can also interpret the results of the table 53 graphically, with the following influences - •POL members w.r.t FOXM1 with FOXM1 −>POL-Q/E2/A1/G2/D1/D2 and •POLR members w.r.t FOXM1 with FOXM1 −>POLR-3K/1C/2G/1B/1A. 2.1.28. FOXM1 - STIL/SF3 In hepatocellular carcinoma (HCC), Zhang et al. [52] showed that an interaction between STIL and FOXM1 regulated the SF3A3 expression. Knockdown of FOXM1 further enhanced the anti-tumor effects of STIL loss on HCC cells in vitro and in vivo, whereas SF3A3 overexpression overturned the impact of STIL loss on HCC cells in vitro and in vivo. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these STIL-SF3 members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these STIL-SF3 members along with FOXM1. Additionally, I also generated rankings of members of STIL-SF3R to see if there was synergy between them and FOXM1. Table 55 shows rankings of these combinations. Followed by this is the unexplored 35 RANKING POL MEMBERS VS FOXM1 RANKING OF POL MEMBERS W.R.TFOXM1 laplace linear rbf laplace linear rbf POLQ - FOXM1 16 11 136 POLE2 - FOXM1 131 801 533 POLA1 - FOXM1 593 771 82 POLG2 - FOXM1 749 1396 831 POLD1 - FOXM1 753 409 611 POLE3 - FOXM1 943 1837 1718 POLD2 - FOXM1 1052 1747 435 POLA2 - FOXM1 1605 1786 1212 POLB - FOXM1 2333 2396 1650 RANKING POLR MEMBERS VS FOXM1 RANKING OF POLR MEMBERS W.R.TFOXM1 laplace linear rbf laplace linear rbf POLR3K - FOXM1 609 868 493 POLR1C - FOXM1 845 850 1218 POLR2G - FOXM1 1147 1074 1739 POLR1B - FOXM1 1401 705 673 POLR1E - FOXM1 1537 1649 1655 POLR1A - FOXM1 1712 1022 1188 POLR3E - FOXM1 2000 1569 2497 POLR2D - FOXM1 2037 1127 1998 POLR2F - FOXM1 2194 2488 1691 POLR1D - FOXM1 2367 2734 2534 POLR3A - FOXM1 2485 2378 2476 POLR2K - FOXM1 2486 2698 2389 Table 53: 2nd order interaction ranking between FOXM1 VS POL members. UNEXPLORED COMBINATORIAL HYPOTHESES POL members w.r.t FOXM1 POL-Q/E2/A1/G2/D1/D2 FOXM1 POLR members w.r.t FOXM1 POLR-3K/1C/2G/1B/1A FOXM1 Table 54: 2nd order combinatorial hypotheses between FOXM1 and POL members. combinatorial hypotheses in table 56 generated from analysis of the ranks in table 55. The table 55 shows rankings of STIL-SF3 members w.r.t FOXM1. SF3A3 - FOXM1 shows low ranking of 1138 (laplace) and 809 (linear). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, STIL, SF3B3 and SF3B5 showed high ranking and might not be synergistically working with FOXM1, before treatment. One can also interpret the results of the table 55 graphically, with the following influences - •SF3 members w.r.t FOXM1 with FOXM1 −>SF3-A3. 2.1.29. FOXM1 - USP/ITG Integrins play critical roles in connecting the extracellular matrix and actin. Liu et al. [53] showed that USP22 was essential in maintaining breast cancer cell stemness by 36 RANKING STIL-SF3 MEMBERS VS FOXM1 RANKING OF STIL-SF3 MEMBERS W.R.TFOXM1 laplace linear rbf STIL - FOXM1 2496 2688 2601 SF3A3 - FOXM1 1138 809 2103 SF3B3 - FOXM1 1310 2447 2338 SF3B5 - FOXM1 2662 2085 2023 Table 55: 2nd order interaction ranking between FOXM1 VS STIL-SF3 members. UNEXPLORED COMBINATORIAL HYPOTHESES STIL-SF3 members w.r.t FOXM1 SF3-A3 FOXM1 Table 56: 2nd order combinatorial hypotheses between FOXM1 and STIL-SF3 members. promoting the transcription of integrin β1 (ITGB1). USP22 functions as a deubiquitinase to protect the proteasomal degradation of FOXM1, a transcription factor for tumoral ITGB1 gene transcription. Immunohistochemistry staining detected a positive correlation among USP22, FOXM1, and integrin β1 in human breast cancers. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC-1922159, these USPITG members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these USP-ITG members along with FOXM1. Table 57 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 58 generated from analysis of the ranks in table 57. The table 57 shows rankings of USP-ITG members w.r.t FOXM1. USP13 - FOXM1 shows low ranking of 485 (laplace), 153 (linear) and 70 (rbf). USP28 - FOXM1 shows low ranking of 1080 (linear) and 1354 (rbf). ITGA9 - FOXM1 shows low ranking of 197 (laplace), 338 (linear) and 496 (rbf). ITGB3BP - FOXM1 shows low ranking of 373 (laplace), 658 (linear) and 374 (rbf). ITGAE - FOXM1 shows low ranking of 1334 (laplace), and 1113 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, USP1, USP10, USP36 and USP39 showed high ranking and might not be synergistically working with FOXM1, before treatment. One can also interpret the results of the table 57 graphically, with the following influences - •USP members w.r.t FOXM1 with FOXM1 −>USP-13/28 and •ITG 37 RANKING USP MEMBERS VS FOXM1 RANKING OF USP MEMBERS W.R.TFOXM1 laplace linear rbf USP13 - FOXM1 485 153 70 USP28 - FOXM1 1741 1080 1354 USP1 - FOXM1 2035 1595 2525 USP10 - FOXM1 2049 2695 2734 USP36 - FOXM1 2463 2363 2084 USP39 - FOXM1 2550 2720 2728 RANKING ITG MEMBERS VS FOXM1 RANKING OF ITG MEMBERS W.R.TFOXM1 laplace linear rbf ITGA9 - FOXM1 197 338 496 ITGB3BP - FOXM1 373 658 374 ITGAE - FOXM1 1334 1691 1113 Table 57: 2nd order interaction ranking between FOXM1 VS USP-ITG members. members w.r.t FOXM1 with FOXM1 −>ITG-A9/B3BP/AE. UNEXPLORED COMBINATORIAL HYPOTHESES USP members w.r.t FOXM1 USP-13/28 FOXM1 ITG members w.r.t FOXM1 ITG-A9/B3BP/AE FOXM1 Table 58: 2nd order combinatorial hypotheses between FOXM1 and USP-ITG members. 38 2.1.30. FOXM1 - ZIC/UBE2 In ccRCC, Lv et al. [54] ZIC2 was upregulated while knockdown of ZIC2 resulted in reduced cell proliferation, invasion, migration, induction of G2/M phase arrest, and reduced tumor formation and lung metastasis in nude mice. Further, hypomethylation and high H3K4Me3 in the promoter region, and positive transcriptional regulation by FOXM1, caused high expression of ZIC2. This ZIC2 transcriptase-positively regulated UBE2C and activated AKT/mTOR signaling pathway to promote tumor malignant progression. These experimental studies confirm the existence of synergy between the above involved factors with FOXM1. In colorectal cancer cells treated with ETC1922159, these ZIC-UBE2 members, and FOXM1, were found to be down regulated and their regulation was recorded independently. I was able to rank 2nd order combination of these ZIC-UBE2 members along with FOXM1. Table 59 shows rankings of these combinations. Followed by this is the unexplored combinatorial hypotheses in table 60 generated from analysis of the ranks in table 59. The table 59 shows rankings of ZIC-UBE2-ITG members w.r.t FOXM1. ZIC2 - FOXM1 shows low ranking of 1109 (laplace) and 260 (rbf). UBE2C - FOXM1 shows low ranking of 15 (laplace), 97 (linear) and 278 (rbf). UBE2T - FOXM1 shows low ranking of 418 (laplace), 83 (linear) and 393 (rbf). UBE2S - FOXM1 shows low ranking of 917 (laplace) and 989 (linear). UBE2G2 - FOXM1 shows low ranking of 1242 (laplace) and 714 (rbf). These rankings point to the synergy existing between the two components, which have been down regulated after the drug treatment. Further, ZIC5 showed high ranking and might not be synergistically working with FOXM1, before treatment. One can also interpret the results of the table 59 graphically, with the following influences - •ZIC members w.r.t FOXM1 with FOXM1 −>ZIC-2 and •ITG members w.r.t FOXM1 with FOXM1 −>UBE2-C/T/S/G2. 3. Conclusion Presented here are a range of multiple synergistic FOXM1 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 FOXM1-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 39 RANKING ZIC MEMBERS VS FOXM1 RANKING OF ZIC MEMBERS W.R.TFOXM1 laplace linear rbf ZIC2 - FOXM1 1109 1993 260 ZIC5 - FOXM1 1929 2588 1647 RANKING UBE2 MEMBERS VS FOXM1 RANKING OF UBE2 MEMBERS W.R.TFOXM1 laplace linear rbf UBE2C - FOXM1 15 97 278 UBE2T - FOXM1 418 83 393 UBE2S - FOXM1 917 989 1875 UBE2G2 - FOXM1 1242 1521 714 Table 59: 2nd order interaction ranking between FOXM1 VS ZIC-UBE2 members. UNEXPLORED COMBINATORIAL HYPOTHESES ZIC members w.r.t FOXM1 ZIC-2 FOXM1 UBE2 members w.r.t FOXM1 UBE2-C/T/S/G2 FOXM1 Table 60: 2nd order combinatorial hypotheses between FOXM1 and ZIC-UBE2-ITG members. Acknowledgements Special thanks to Mrs. Rita Sinha and late 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. [55]. 40 4. References References [1] H. Ye, T. F. Kelly, U. Samadani, L. Lim, S. Rubio, D. G. Overdier, K. A. Roebuck, R. H. Costa, Hepatocyte nuclear factor 3/fork head homolog 11 is expressed in proliferating epithelial and mesenchymal cells of embryonic and adult tissues, Molecular and cellular biology 17 (1997) 1626–1641. [2] M. Castaneda, P. d. Hollander, S. A. Mani, Forkhead box transcription factors: double-edged swords in cancer, Cancer research 82 (2022) 2057–2065. [3] I. Wierstra, J. Alves, Foxm1, a typical proliferation-associated transcription factor (2007). [4] C. Yang, H. Chen, G. Tan, W. Gao, L. Cheng, X. Jiang, L. Yu, Y. Tan, Foxm1 promotes the epithelial to mesenchymal transition by stimulating the transcription of slug in human breast cancer, Cancer letters 340 (2013) 104–112. [5] R. E. Francis, S. S. Myatt, J. Krol, J. Hartman, B. Peck, U. B. McGovern, J. Wang, S. K. Guest, A. Filipovic, O. Gojis, et al., Foxm1 is a downstream target and marker of her2 overexpression in breast cancer, International journal of oncology 35 (2009) 57–68. [6] Y. Liu, Z. Gong, L. Sun, X. Li, Foxm1 and androgen receptor co-regulate cdc6 gene transcription and dna replication in prostate cancer cells, Biochimica et Biophysica Acta (BBA)-Gene Regulatory Mechanisms 1839 (2014) 297–305. [7] J. Cui, M. Shi, D. Xie, D. Wei, Z. Jia, S. Zheng, Y. Gao, S. Huang, K. Xie, Foxm1 promotes the warburg effect and pancreatic cancer progression via transactivation of ldha expression, Clinical Cancer Research 20 (2014) 2595–2606. [8] J. Zeng, L. Wang, Q. Li, W. Li, M. Bjorkholm, J. Jia, D. Xu, Foxm1 is up-regulated in gastric cancer and its inhibition leads to cellular senescence, partially dependent on p27kip1, The Journal of Pathology: A Journal of the Pathological Society of Great Britain and Ireland 218 (2009) 419–427. [9] 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. [10] 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. [11] 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. [12] H. Wang, S. Han, J. Xiao, X. Fu, W. Chen, D. Zhuo, Otud7b knockdown inhibits the proliferation and stemness of breast cancer cells by destabilizing foxm1, Oncology Letters 27 (2024) 1–12. [13] Q. Tang, C. Liu, S. Zhang, L. He, Y. Liu, J. Wang, X. Zhao, X. Li, Foxm1 increases htert protein stability and indicates poor prognosis in gastric cancer, Neoplasia 36 (2023) 100863. [14] C. J. Barger, L. Chee, M. Albahrani, C. Munoz-Trujillo, L. Boghean, C. Branick, K. Odunsi, R. Drapkin, L. Zou, A. R. Karpf, Co-regulation and function of foxm1/rhno1 bidirectional genes in cancer, Elife 10 (2021) e55070. [15] J. Liu, J. Li, K. Wang, H. Liu, J. Sun, X. Zhao, Y. Yu, Y. Qiao, Y. Wu, X. Zhang, et al., Aberrantly high activation of a foxm1–stmn1 axis contributes to progression and tumorigenesis in foxm1-driven cancers, Signal transduction and targeted therapy 6 (2021) 42. [16] H. Xie, N. Miao, D. Xu, Z. Zhou, J. Ni, F. Yin, Y. Wang, Q. Cheng, P. Chen, J. Li, et al., Foxm1 promotes wnt/βcatenin pathway activation and renal fibrosis via transcriptionally regulating multi-wnts expressions, Journal of Cellular and Molecular Medicine 25 (2021) 1958–1971. [17] Y. Wang, H. Wang, Z. Yan, G. Li, G. Hu, H. Zhang, D. Huang, Y. Wang, X. Zhang, Y. Yan, et al., The critical role of dysregulated hh-foxm1-tpx2 signaling in human hepatocellular carcinoma cell proliferation, Cell Communication and Signaling 18 (2020) 1–18. 41