%Version 3.1 December 2024 % See section 11 of the User Manual for version history % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %% %% %% Please do not use \input{...} to include other tex files. %% %% Submit your LaTeX manuscript as one .tex document. %% %% %% %% All additional figures and files should be attached %% %% separately and not embedded in the \TeX\ document itself. %% %% %% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %%\documentclass[referee,sn-basic]{sn-jnl}% referee option is meant for double line spacing %%=======================================================%% %% to print line numbers in the margin use lineno option %% %%=======================================================%% %%\documentclass[lineno,pdflatex,sn-basic]{sn-jnl}% Basic Springer Nature Reference Style/Chemistry Reference Style %%=========================================================================================%% %% the documentclass is set to pdflatex as default. You can delete it if not appropriate. %% %%=========================================================================================%% %%\documentclass[sn-basic]{sn-jnl}% Basic Springer Nature Reference Style/Chemistry Reference Style %%Note: the following reference styles support Namedate and Numbered referencing. By default the style follows the most common style. To switch between the options you can add or remove “Numbered” in the optional parenthesis. %%The option is available for: sn-basic.bst, sn-chicago.bst% %%\documentclass[pdflatex,sn-nature]{sn-jnl}% Style for submissions to Nature Portfolio journals \documentclass[pdflatex,sn-basic]{sn-jnl}% Basic Springer Nature Reference Style/Chemistry Reference Style %%\documentclass[pdflatex,sn-mathphys-num]{sn-jnl}% Math and Physical Sciences Numbered Reference Style %%\documentclass[pdflatex,sn-mathphys-ay]{sn-jnl}% Math and Physical Sciences Author Year Reference Style %%\documentclass[pdflatex,sn-aps]{sn-jnl}% American Physical Society (APS) Reference Style %%\documentclass[pdflatex,sn-vancouver-num]{sn-jnl}% Vancouver Numbered Reference Style %%\documentclass[pdflatex,sn-vancouver-ay]{sn-jnl}% Vancouver Author Year Reference Style %%\documentclass[pdflatex,sn-apa]{sn-jnl}% APA Reference Style %%\documentclass[pdflatex,sn-chicago]{sn-jnl}% Chicago-based Humanities Reference Style %%%% Standard Packages %% \usepackage{graphicx}% \usepackage{multirow}% \usepackage{amsmath,amssymb,amsfonts}% \usepackage{amsthm}% \usepackage{mathrsfs}% \usepackage[title]{appendix}% \usepackage{xcolor}% \usepackage{textcomp}% \usepackage{manyfoot}% \usepackage{booktabs}% \usepackage{algorithm}% \usepackage{algorithmicx}% \usepackage{algpseudocode}% \usepackage{listings}% \usepackage[acronym]{glossaries} \usepackage{todonotes} %%%% \newacronym{SAR}{SAR}{structure-activity relationship} \newacronym{GPCR}{GPCR}{G-protein-coupled receptor} \newacronym{LBVS}{LBVS}{ligand-based virtual screening} \newacronym{SBVS}{SBVS}{structure-based virtual screening} \newacronym{ECFP}{ECFP}{extended-connectivity fingerprints} \newacronym{QSAR}{QSAR}{quantitative structure-activity relationship} \newacronym{EF1}{EF@1\%}{enrichment factor at 1\%} \newacronym{ROC}{ROC}{receiver operating characteristic} \newacronym{AUC}{AUC}{area under the curve} \newacronym{PR}{PR}{precision-recall} \newacronym{RIE}{RIE}{robust initial enhancement} \newacronym{BEDROC}{BEDROC}{Boltzmann-enhanced discrimination of ROC} \newacronym{MF}{MF}{molecular function} \newacronym{RS}{RS}{reference set} \newacronym{UMAP}{UMAP}{uniform manifold approximation and projection} \newacronym{PCA}{PCA}{principal component analysis} \newacronym{SVM}{SVM}{support vector machine} \newacronym{1-NN}{1-NN}{1-nearest-neighbor} %%%% %% as per the requirement new theorem styles can be included as shown below \theoremstyle{thmstyleone}% \newtheorem{theorem}{Theorem}% meant for continuous numbers %%\newtheorem{theorem}{Theorem}[section]% meant for sectionwise numbers %% optional argument [theorem] produces theorem numbering sequence instead of independent numbers for Proposition \newtheorem{proposition}[theorem]{Proposition}% %%\newtheorem{proposition}{Proposition}% to get separate numbers for theorem and proposition etc. \theoremstyle{thmstyletwo}% \newtheorem{example}{Example}% \newtheorem{remark}{Remark}% \theoremstyle{thmstylethree}% \newtheorem{definition}{Definition}% \raggedbottom %%\unnumbered% uncomment this for unnumbered level heads \begin{document} \title{Supplementary Material for:\\ ``Molecular Function-guided Unsupervised Similarity-based Enrichment for Low-Data Virtual Screening''} \author*[1,2]{\fnm{Alexander} \sur{Hagg}}\email{alexander.hagg@h-brs.de} \author[1,2,3]{\fnm{Dirk} \sur{Reith}}\email{dirk.reith@h-brs.de} \author[2,4,5]{\fnm{Matthias} \sur{Preller}}\email{matthias.preller@h-brs.de} \author[2,6]{\fnm{Karl N.} \sur{Kirschner}}\email{karl.kirschner@h-brs.de} \affil*[1]{\orgdiv{Department of Engineering and Communication}, \orgname{University of Applied Sciences Bonn-Rhein-Sieg}, \orgaddress{\street{Grantham-Allee 20}, \city{Sankt Augustin}, \postcode{53757}, \country{Germany}}} \affil*[2]{\orgdiv{Institute of Technology, Resource and Energy-efficient Engineering}, \orgname{University of Applied Sciences Bonn-Rhein-Sieg}, \orgaddress{\street{Grantham-Allee 20}, \city{Sankt Augustin}, \postcode{53757}, \country{Germany}}} \affil[3]{\orgname{Fraunhofer Institute for Algorithms and Scientific Computing}, \orgaddress{\street{Schloss Birlinghoven}, \city{Sankt Augustin}, \postcode{53754}, \country{Germany}}} \affil[4]{\orgdiv{Department of Natural Sciences}, \orgname{University of Applied Sciences Bonn-Rhein-Sieg}, \orgaddress{\street{Von-Liebig-Straße 20}, \city{Rheinbach}, \postcode{53359}, \country{Germany}}} \affil[5]{\orgdiv{Institute for Functional Gene Analytics}, \orgname{University of Applied Sciences Bonn-Rhein-Sieg}, \orgaddress{\street{Von-Liebig-Straße 20}, \city{Rheinbach}, \postcode{53359}, \country{Germany}}} \affil*[6]{\orgdiv{Department of Computer Science}, \orgname{University of Applied Sciences Bonn-Rhein-Sieg}, \orgaddress{\street{Grantham-Allee 20}, \city{Sankt Augustin}, \postcode{53757}, \country{Germany}}} \maketitle \section{Extended Related Work}\label{SI:related_work} This section provides extended context on virtual screening metrics, baseline screening performance, deep learning architectures, and dimensionality reduction methodologies as documented in recent literature. \subsection{Metrics and Benchmarks in Virtual Screening} Quantifying the ability of virtual screening methods to prioritize active compounds -- known as enrichment -- is fundamental to evaluating their utility in early-stage drug discovery. While the literature presents a diverse array of evaluation metrics and benchmark datasets \cite{mcgaughey2007comparison, sciabola2022critical}, which often complicate direct cross-study comparisons, robust metrics such as the \gls{EF1} have become standard for assessing early retrieval success. Since the datasets used in previous studies differ from the molecular function-guided configuration employed here, the performance figures below provide a general context for indirect comparison rather than a direct benchmark. A critical analysis of structure-based virtual screening (\gls{SBVS}) benchmarks highlight that established resources such as the Directory of Useful Decoys (DUD) and its enhanced version (DUD-E) \cite{mysinger2012directory} exhibit severe analogue and decoy biases \cite{rohrer2009maximum}. These systematic biases can inflate performance estimates by 2--5 times relative to more recent, unbiased benchmarks such as LIT-PCBA \cite{tran2020lit}. \subsection{Advanced Virtual Screening Scoring Functions and Ensembles} When experimental target data is known, specialized scoring functions have reported high enrichments on standardized benchmarks. For example, the Similarity of Interaction Energy VEctor Score (SIEVE-Score), which incorporates molecular docking within its workflow, achieved a median \gls{EF1} of 44.9 (10\%/90\% percentiles [27.3, 60.0]) on a subset of DUD-E targets, substantially outperforming classical Glide docking \cite{yasuo2019improved}. In comparison, the machine-learned RF-Score on the same subset yielded median \gls{EF1} values of 18.7 (10\%/90\% percentiles [4.1, 43.0]) and 38.6 (10\%/90\% percentiles [13.8, 58.9]) depending on training configuration \cite{yasuo2019improved}. Hybrid screening approaches have also demonstrated elevated performance. A protocol combining ensemble docking with electrostatic comparisons (e-Sim) \cite{cleves2019electrostatic} achieved a median \gls{EF1} of 46.1 across a 96-protein DUD-E subset \cite{cleves2020structure}. Furthermore, the ensemble learning method ENS-VS -- which integrates support vector machines (\glspl{SVM}), a decision tree, and a Fisher linear discriminant -- achieved a median \gls{EF1} of 57.0 (10\%/90\% percentiles [51.2, 63.9]) on DUD-E, compared to AutoDock Vina alone which reached 4.7 (10\%/90\% percentiles [3.3, 17.7]) \cite{li2020improved}. When evaluated on the DEKOIS dataset \cite{vogel2011dekois}, ENS-VS yielded a median \gls{EF1} of 29.5 (10\%/90\% percentiles [21.2, 37.9]), compared to Vina (2.4), Glide (9.5), SIEVE-Score (28.1), and RF-ScoreVS\_v3\_vina (16.6) \cite{li2020improved}. \subsection{Deep Learning Representation Models} Modern deep learning architectures focus on navigating high-dimensional molecular descriptor spaces or generating latent representations of these spaces. Recent Graph Neural Networks (GNNs) combining learned topologies with expert-curated descriptors have been trained on large-scale bioactivity datasets containing between 60,000 and 350,000 compounds \cite{liu2025advancements}. Similarly, Transformer models adapted for chemistry learn fundamental structural rules through pre-training on millions of unlabeled chemical structures \cite{MswahiliJ2024, BanerjeeKRPSKGB2025, UmerNALAUF2025}. Models like ChemBERTa \cite{chithrananda2020chemberta} are designed for fine-tuning on diverse downstream tasks. Another prominent transformer model, MolBERT \cite{fabian2020molecular}, designed to produce generalized molecular representations, achieved a BEDROC score of 0.344 $\pm$ 0.062 across 69 unique protein datasets \cite{fabian2020molecular}. While computationally powerful, these frameworks generally demand substantial labeled training data and high-performance computing resources, and their internal mechanisms remain highly complex to interpret compared to distance-based metric retrieval. \subsection{Dimensionality Reduction Paradigms in Cheminformatics} Chemical descriptor packages often generate highly multidimensional representations. Dimensionality reduction is critical to compress these spaces, extract latent chemical features, and permit visualization \cite{NguyenH2019}. Linear dimensionality reduction techniques like Principal Component Analysis (\gls{PCA}) generate embeddings by projecting data along directions of maximal variance \cite{oprea2001chemography}. However, because PCA relies entirely on linear transformations, it frequently struggles to maintain local, non-linear neighborhood structures that dictate molecular similarity and biological activity \cite{orlov2025high, lovric2021should}. Conversely, non-linear techniques like t-distributed Stochastic Neighbor Embedding (t-SNE) \cite{van2008visualizing} and Uniform Manifold Approximation and Projection (\gls{UMAP}) \cite{mcinnes2018umap} resolve some of these limitations but utilize different mathematical frameworks. t-SNE focuses heavily on preserving local distances, which often causes global spatial relationships and inter-cluster distances between distinct chemotypes to lose quantitative meaning \cite{kobak2019art}. In contrast, \gls{UMAP} relies on Riemannian geometry and algebraic topology, allowing it to preserve local neighborhoods while maintaining global structure. This balance makes relative cluster positioning more physically interpretable. Furthermore, \gls{UMAP} is computationally faster, scales more effectively to large datasets, and offers standard formulations to project novel, out-of-sample molecules into an existing embedding space without requiring model retraining \cite{becht2019dimensionality}. While manifold learning approaches are routinely utilized for visual representation \cite{probst2020visualization, cihan2022chemplot, orlov2025high} and clustering \cite{Hernandez-HernandezB2023, GuoHB2024, GuoHB2025}, their application as a direct distance-based scoring metric for virtual screening retrieval remains a relatively underexplored paradigm compared to classical fingerprint comparisons or linear projections. \section{Molecular Descriptors} The 1,611 Mordred 2D molecular descriptors used in this work are organised into twelve categories in Tables~\ref{tab:desc1}--\ref{tab:desc1b} and \ref{tab:desc2}--\ref{tab:desc5}. The autocorrelation weight codes used throughout are: $c$ = Gasteiger charge, $dv$ = valence electrons, $d$ = sigma electrons, $s$ = intrinsic state, $Z$ = atomic number, $m$ = atomic mass, $v$ = van-der-Waals volume, $se$ = Sanderson electronegativity, $pe$ = Pauling electronegativity, $are$ = Allred--Rochow electronegativity, $p$ = polarizability, $i$ = first ionization potential. % ----------------------------------------------------------------------- % Table 1a: Constitutional Descriptors % ----------------------------------------------------------------------- \begin{table}[htbp] \centering \caption{Constitutional Descriptors.} \label{tab:desc1} \small \begin{tabular}{p{4.5cm}p{7.5cm}} \hline \textbf{Descriptor(s)} & \textbf{Description} \\ \hline $MW$, $AMW$ & Molecular weight and average molecular weight. \\ $nAtom$, $nHeavyAtom$, $nH$ & Total atom count, heavy atom count, hydrogen count. \\ $nHetero$, $nX$ & Heteroatom and halogen counts. \\ $nB$, $nC$, $nN$, $nO$, $nS$, $nP$ & Per-element atom counts (B, C, N, O, S, P). \\ $nF$, $nCl$, $nBr$, $nI$ & Halogen atom counts (F, Cl, Br, I). \\ $nAromAtom$, $nAromBond$ & Aromatic atom and bond counts. \\ $nSpiro$, $nBridgehead$ & Spiro and bridgehead atom counts. \\ $nAcid$, $nBase$ & Number of acidic and basic functional groups. \\ $nHBAcc$, $nHBDon$ & Hydrogen-bond acceptor and donor counts. \\ $nRot$, $RotRatio$ & Rotatable bond count and ratio. \\ $nBonds$, $nBondsO$, $nBondsS$ & Total bond count, single bonds, and bond-order sum. \\ $nBondsD$, $nBondsT$, $nBondsA$ & Double, triple, and aromatic bond counts. \\ $nBondsM$, $nBondsKS$, $nBondsKD$ & Multiple, Kekulé-single, and Kekulé-double bond counts. \\ $RNCG$, $RPCG$ & Relative negative/positive charge (most charged atom fraction). \\ $FCSP3$, $HybRatio$ & Fraction of $sp^3$ carbons; ratio of $sp^3$ atoms to heavy atoms. \\ $C1SP1$–$C4SP3$ & Carbon counts by hybridization state and substitution ($sp$, $sp^2$, $sp^3$). \\ $fragCpx$, $fMF$ & Fragment complexity and molecular framework fraction. \\ $apol$, $bpol$ & Sum of atomic and bond polarizabilities. \\ $VMcGowan$, $Vabc$, $VAdjMat$ & McGowan characteristic volume, ABC volume, adjacency-matrix volume. \\ $LabuteASA$ & Labute approximate surface area. \\ \hline \end{tabular} \end{table} % ----------------------------------------------------------------------- % Table 1b: Graph-Theoretical & Physicochemical Descriptors % ----------------------------------------------------------------------- \begin{table}[htbp] \centering \caption{Graph-Theoretical and Physicochemical Descriptors.} \label{tab:desc1b} \small \begin{tabular}{p{4.5cm}p{7.5cm}} \hline \textbf{Descriptor(s)} & \textbf{Description} \\ \hline $ABC$, $ABCGG$ & Atom-Bond Connectivity index and Graovac--Ghorbani variant. \\ $BalabanJ$ & Balaban's connectivity (J) index. \\ $WPath$, $WPol$ & Wiener path index and Wiener polarity index. \\ $Zagreb1$, $Zagreb2$, $mZagreb1$, $mZagreb2$ & First and second Zagreb indices and their modified variants. \\ $BertzCT$ & Bertz complexity index. \\ $ECIndex$ & Eccentric connectivity index. \\ $DetourIndex$ & Detour index (longest-path sum). \\ $Diameter$, $Radius$ & Topological diameter and radius. \\ $TopoShapeIndex$, $PetitjeanIndex$ & Topological and Petitjean shape indices. \\ $SLogP$, $SMR$ & Wildman--Crippen $\log P$ and molar refractivity estimates. \\ $FilterItLogS$ & Filter-it predicted aqueous solubility ($\log S$). \\ $TopoPSA$, $TopoPSA(NO)$ & Topological polar surface area (all heteroatoms; N/O only). \\ \hline \end{tabular} \end{table} % ----------------------------------------------------------------------- % Table 2: Autocorrelation Descriptors % ----------------------------------------------------------------------- \begin{table}[htbp] \centering \caption{Autocorrelation Descriptors. Weight codes $w \in \{c, dv, d, s, Z, m, v, se, pe, are, p, i\}$ (see text); lag $k$ ranges as noted.} \label{tab:desc2} \small \begin{tabular}{lp{4.5cm}p{7cm}} \hline \textbf{Category} & \textbf{Descriptor(s)} & \textbf{Description} \\ \hline \multirow{8}{*}{\textbf{Autocorrelation}} & $ATS_k w$ & Moreau--Broto autocorrelation, lag $k=0$--$8$; weights $w \in \{dv, d, s, Z, m, v, se, pe, are, p, i\}$. \\ & $AATS_k w$ & Average Moreau--Broto autocorrelation, lag $k=0$--$8$; same weights as $ATS$. \\ & $ATSC_k w$ & Centred Moreau--Broto autocorrelation, lag $k=0$--$8$; weights $w \in \{c, dv, d, s, Z, m, v, se, pe, are, p, i\}$. \\ & $AATSC_k w$ & Average centred Moreau--Broto autocorrelation, lag $k=0$--$8$; same weights as $ATSC$. \\ & $MATS_k w$ & Moran autocorrelation index, lag $k=1$--$8$; weights $w \in \{c, dv, d, s, Z, m, v, se, pe, are, p, i\}$. \\ & $GATS_k w$ & Geary autocorrelation index, lag $k=1$--$8$; same weights as $MATS$. \\ \hline \multirow{2}{*}{\textbf{BCUT}} & $BCUTw\text{-}1h$, $BCUTw\text{-}1l$ & Highest and lowest eigenvalues of the Burden matrix weighted by property $w \in \{c, dv, d, s, Z, m, v, se, pe, are, p, i\}$ (24 descriptors total). \\ \hline \end{tabular} \end{table} % ----------------------------------------------------------------------- % Table 3: Matrix Spectral, Connectivity & E-State % ----------------------------------------------------------------------- \begin{table}[htbp] \centering \caption{Matrix Spectral, Connectivity, and E-State Descriptors.} \label{tab:desc3} \small \begin{tabular}{lp{4.5cm}p{7cm}} \hline \textbf{Category} & \textbf{Descriptor(s)} & \textbf{Description} \\ \hline \multirow{10}{*}{\textbf{Matrix Spectral}} & $SpAbs\_A$, $SpMax\_A$, $SpDiam\_A$ & Spectral absolute value, maximum eigenvalue, and spectral diameter of the adjacency matrix ($A$). \\ & $SpAD\_A$, $SpMAD\_A$, $LogEE\_A$ & Spectral absolute deviation, mean absolute deviation, and log of Estrada index for $A$. \\ & $VE1\_A$--$VE3\_A$, $VR1\_A$--$VR3\_A$ & Eigenvector-based and Randić-type spectral coefficients of $A$. \\ & $SpAbs\_D$, $SpMax\_D$, $\ldots$, $VR3\_D$ & Analogous spectral descriptors for the topological distance matrix ($D$). \\ & $SpAbs\_Dt$, $SpMax\_Dt$, $\ldots$, $VR3\_Dt$, $SM1\_Dt$ & Spectral descriptors for the detour matrix ($Dt$), including $SM1$ (sum of eigenvalues). \\ & $SpAbs\_Dz w$, $\ldots$, $VR3\_Dz w$, $SM1\_Dz w$ & Spectral descriptors for property-weighted distance matrices ($Dz$), $w \in \{Z, m, v, se, pe, are, p, i\}$ (8 matrices, 13 stats each = 104 descriptors). \\ & $SZ$, $Sm$, $Sv$, $Sse$, $Spe$, $Sare$, $Sp$, $Si$ & Sum of atomic property values over all heavy atoms ($Z$, $m$, $v$, $se$, $pe$, $are$, $p$, $i$). \\ & $MZ$, $Mm$, $Mv$, $Mse$, $Mpe$, $Mare$, $Mp$, $Mi$ & Mean of atomic property values. \\ \hline \multirow{5}{*}{\textbf{Connectivity}} & $Xp\text{-}k d$, $Xp\text{-}k dv$ & Path chi indices, orders $k=0$--$7$, for $\sigma$-electron ($d$) and valence-electron ($dv$) weights. \\ & $AXp\text{-}k d$, $AXp\text{-}k dv$ & Average path chi indices, orders $k=0$--$7$. \\ & $Xc\text{-}k d$, $Xc\text{-}k dv$ & Cluster chi indices, orders $k=3$--$6$. \\ & $Xpc\text{-}k d$, $Xpc\text{-}k dv$ & Path-cluster chi indices, orders $k=4$--$6$. \\ & $Xch\text{-}k d$, $Xch\text{-}k dv$ & Chain chi indices, orders $k=3$--$7$. \\ & $Kier1$, $Kier2$, $Kier3$ & Kier shape indices ($\kappa_1$, $\kappa_2$, $\kappa_3$). \\ \hline \multirow{5}{*}{\textbf{E-State}} & $Ns XY$ & Count of atom type $XY$ (e.g., $NsCH3$, $NaaNH$, $NsOH$); covers C, N, O, S, P, Si, Ge, As, Se, Sn, Pb, and halogen environments. \\ & $Ss XY$ & E-State index sum over atom type $XY$. \\ & $MAXs XY$ & Maximum E-State value over atom type $XY$. \\ & $MINs XY$ & Minimum E-State value over atom type $XY$. \\ & $EState\_VSA1$--$10$ & Surface area contributions binned by E-State value. \\ & $VSA\_EState1$--$9$ & E-State contributions binned by surface area. \\ \hline \end{tabular} \end{table} % ----------------------------------------------------------------------- % Table 4: Surface Area, Ring & Walk, Information Indices % ----------------------------------------------------------------------- \begin{table}[htbp] \centering \caption{Surface Area, Ring/Walk, and Information-Theoretic Descriptors.} \label{tab:desc4} \small \begin{tabular}{lp{4.5cm}p{7cm}} \hline \textbf{Category} & \textbf{Descriptor(s)} & \textbf{Description} \\ \hline \multirow{4}{*}{\textbf{Surface Area}} & $SlogP\_VSA1$--$11$ & Surface area contributions binned by Wildman--Crippen $\log P$ values. \\ & $SMR\_VSA1$--$9$ & Surface area contributions binned by molar refractivity. \\ & $PEOE\_VSA1$--$13$ & Surface area contributions binned by Gasteiger partial charges. \\ \hline \multirow{6}{*}{\textbf{Rings \& Walks}} & $nRing$, $n3Ring$--$n12Ring$, $nG12Ring$ & Total ring count and ring-size counts (3--12 members; $nG12Ring$ for rings $>12$). \\ & $nHRing$, $n3HRing$--$nG12HRing$ & Heteroatom-containing ring counts by size. \\ & $naRing$, $n3aRing$--$nG12aRing$ & Aromatic ring counts by size. \\ & $naHRing$, $n3aHRing$--$nG12aHRing$ & Aromatic heteroatom ring counts by size. \\ & $nARing$, $nAHRing$ and size variants & Aliphatic ring and aliphatic heteroatom ring counts. \\ & $nFRing$, $nFHRing$, $nFaRing$, $nFaHRing$, $nFARing$, $nFAHRing$ and size variants & Fused-ring counts (fused hetero, aromatic, aliphatic variations) by ring size. \\ & $MWC01$--$MWC10$, $TMWC10$ & Molecular Walk Counts of lengths 1--10 and total. \\ & $SRW02$--$SRW10$, $TSRW10$ & Self-Returning Walk counts of lengths 2--10 and total. \\ \hline \multirow{5}{*}{\textbf{Information}} & $IC0$--$IC5$ & Information content indices at graph distance 0--5. \\ & $TIC0$--$TIC5$ & Total information content indices. \\ & $SIC0$--$SIC5$ & Structural information content indices. \\ & $BIC0$--$BIC5$ & Bonding information content indices. \\ & $CIC0$--$CIC5$ & Complementary information content indices. \\ & $MIC0$--$MIC5$ & Modified information content indices. \\ & $ZMIC0$--$ZMIC5$ & Atomic number-weighted modified information content. \\ \hline \multirow{3}{*}{\textbf{Distance Codes}} & $MID$, $AMID$ & Mean information-theoretic distance and its average. \\ & $MID\_h$, $AMID\_h$ & $MID$ / $AMID$ restricted to hydrogen-containing paths. \\ & $MID\_C$, $AMID\_C$; $MID\_N$, $AMID\_N$; $MID\_O$, $AMID\_O$; $MID\_X$, $AMID\_X$ & Element-resolved mean distances (C, N, O, halogen). \\ \hline \end{tabular} \end{table} % ----------------------------------------------------------------------- % Table 5: Topological Charge, Walk/Path Counts, ETA, Physicochemical % ----------------------------------------------------------------------- \begin{table}[htbp] \centering \caption{Topological Charge, Walk/Path Counts, and Topochemical (ETA) Descriptors.} \label{tab:desc5} \small \begin{tabular}{lp{4.5cm}p{7cm}} \hline \textbf{Category} & \textbf{Descriptor(s)} & \textbf{Description} \\ \hline \multirow{4}{*}{\textbf{Topological Charge}} & $GGI1$--$GGI10$ & Galvez topological charge indices of order 1--10. \\ & $JGI1$--$JGI10$ & Mean topological charge indices of order 1--10. \\ & $JGT10$ & Total topological charge index (sum over orders 1--10). \\ \hline \multirow{4}{*}{\textbf{Path/Walk Counts}} & $MPC2$--$MPC10$, $TMPC10$ & Molecular path counts of length 2--10 and total. \\ & $piPC1$--$piPC10$, $TpiPC10$ & $\pi$-path counts of length 1--10 and total. \\ & $MDEC\text{-}ij$ & Molecular distance-edge carbon descriptors ($ij \in \{11,12,13,14,22,23,24,33,34,44\}$). \\ & $MDEO\text{-}ij$ & Molecular distance-edge oxygen descriptors ($ij \in \{11,12,22\}$). \\ & $MDEN\text{-}ij$ & Molecular distance-edge nitrogen descriptors ($ij \in \{11,12,13,22,23,33\}$). \\ \hline \multirow{8}{*}{\textbf{Topochemical (ETA)}} & $ETA\_alpha$, $AETA\_alpha$ & ETA alpha (composite atom contribution) and average. \\ & $ETA\_shape\_p$, $ETA\_shape\_y$, $ETA\_shape\_x$ & ETA shape descriptors (parabolic, Y-shaped, X-shaped molecular profiles). \\ & $ETA\_beta$, $AETA\_beta$, $ETA\_beta\_s$, $AETA\_beta\_s$ & ETA beta (total and sigma-only) and averages. \\ & $ETA\_beta\_ns$, $AETA\_beta\_ns$, $ETA\_beta\_ns\_d$, $AETA\_beta\_ns\_d$ & ETA beta non-sigma contributions and their differences. \\ & $ETA\_eta$, $AETA\_eta$, $ETA\_eta\_L$, $AETA\_eta\_L$ & ETA eta indices (local and average). \\ & $ETA\_eta\_R$, $AETA\_eta\_R$, $ETA\_eta\_RL$, $AETA\_eta\_RL$ & ETA eta R-branch indices and averages. \\ & $ETA\_eta\_F$, $AETA\_eta\_F$, $ETA\_eta\_FL$, $AETA\_eta\_FL$ & ETA eta F indices and averages. \\ & $ETA\_eta\_B$, $AETA\_eta\_B$, $ETA\_eta\_BR$, $AETA\_eta\_BR$ & ETA eta B indices and averages. \\ & $ETA\_dAlpha\_A$, $ETA\_dAlpha\_B$ & Delta-alpha ETA descriptors (A and B variants). \\ & $ETA\_epsilon\_1$--$5$ & ETA epsilon descriptors (five variants). \\ & $ETA\_dEpsilon\_A$--$D$ & Delta-epsilon ETA descriptors. \\ & $ETA\_dBeta$, $AETA\_dBeta$ & Delta-beta and average delta-beta. \\ & $ETA\_psi\_1$, $ETA\_dPsi\_A$, $ETA\_dPsi\_B$ & ETA psi and delta-psi descriptors. \\ \hline \end{tabular} \end{table} % USED FEATURES: %ABC,ABCGG,nAcid,nBase,SpAbs_A,SpMax_A,SpDiam_A,SpAD_A,SpMAD_A,LogEE_A,VE1_A,VE2_A,VE3_A,VR1_A,VR2_A,VR3_A,nAromAtom,nAromBond,nAtom,nHeavyAtom,nSpiro,nBridgehead,nHetero,nH,nB,nC,nN,nO,nS,nP,nF,nCl,nBr,nI,nX,ATS0dv,ATS1dv,ATS2dv,ATS3dv,ATS4dv,ATS5dv,ATS6dv,ATS7dv,ATS8dv,ATS0d,ATS1d,ATS2d,ATS3d,ATS4d,ATS5d,ATS6d,ATS7d,ATS8d,ATS0s,ATS1s,ATS2s,ATS3s,ATS4s,ATS5s,ATS6s,ATS7s,ATS8s,ATS0Z,ATS1Z,ATS2Z,ATS3Z,ATS4Z,ATS5Z,ATS6Z,ATS7Z,ATS8Z,ATS0m,ATS1m,ATS2m,ATS3m,ATS4m,ATS5m,ATS6m,ATS7m,ATS8m,ATS0v,ATS1v,ATS2v,ATS3v,ATS4v,ATS5v,ATS6v,ATS7v,ATS8v,ATS0se,ATS1se,ATS2se,ATS3se,ATS4se,ATS5se,ATS6se,ATS7se,ATS8se,ATS0pe,ATS1pe,ATS2pe,ATS3pe,ATS4pe,ATS5pe,ATS6pe,ATS7pe,ATS8pe,ATS0are,ATS1are,ATS2are,ATS3are,ATS4are,ATS5are,ATS6are,ATS7are,ATS8are,ATS0p,ATS1p,ATS2p,ATS3p,ATS4p,ATS5p,ATS6p,ATS7p,ATS8p,ATS0i,ATS1i,ATS2i,ATS3i,ATS4i,ATS5i,ATS6i,ATS7i,ATS8i,AATS0dv,AATS1dv,AATS2dv,AATS3dv,AATS4dv,AATS5dv,AATS6dv,AATS7dv,AATS8dv,AATS0d,AATS1d,AATS2d,AATS3d,AATS4d,AATS5d,AATS6d,AATS7d,AATS8d,AATS0s,AATS1s,AATS2s,AATS3s,AATS4s,AATS5s,AATS6s,AATS7s,AATS8s,AATS0Z,AATS1Z,AATS2Z,AATS3Z,AATS4Z,AATS5Z,AATS6Z,AATS7Z,AATS8Z,AATS0m,AATS1m,AATS2m,AATS3m,AATS4m,AATS5m,AATS6m,AATS7m,AATS8m,AATS0v,AATS1v,AATS2v,AATS3v,AATS4v,AATS5v,AATS6v,AATS7v,AATS8v,AATS0se,AATS1se,AATS2se,AATS3se,AATS4se,AATS5se,AATS6se,AATS7se,AATS8se,AATS0pe,AATS1pe,AATS2pe,AATS3pe,AATS4pe,AATS5pe,AATS6pe,AATS7pe,AATS8pe,AATS0are,AATS1are,AATS2are,AATS3are,AATS4are,AATS5are,AATS6are,AATS7are,AATS8are,AATS0p,AATS1p,AATS2p,AATS3p,AATS4p,AATS5p,AATS6p,AATS7p,AATS8p,AATS0i,AATS1i,AATS2i,AATS3i,AATS4i,AATS5i,AATS6i,AATS7i,AATS8i,ATSC0c,ATSC1c,ATSC2c,ATSC3c,ATSC4c,ATSC5c,ATSC6c,ATSC7c,ATSC8c,ATSC0dv,ATSC1dv,ATSC2dv,ATSC3dv,ATSC4dv,ATSC5dv,ATSC6dv,ATSC7dv,ATSC8dv,ATSC0d,ATSC1d,ATSC2d,ATSC3d,ATSC4d,ATSC5d,ATSC6d,ATSC7d,ATSC8d,ATSC0s,ATSC1s,ATSC2s,ATSC3s,ATSC4s,ATSC5s,ATSC6s,ATSC7s,ATSC8s,ATSC0Z,ATSC1Z,ATSC2Z,ATSC3Z,ATSC4Z,ATSC5Z,ATSC6Z,ATSC7Z,ATSC8Z,ATSC0m,ATSC1m,ATSC2m,ATSC3m,ATSC4m,ATSC5m,ATSC6m,ATSC7m,ATSC8m,ATSC0v,ATSC1v,ATSC2v,ATSC3v,ATSC4v,ATSC5v,ATSC6v,ATSC7v,ATSC8v,ATSC0se,ATSC1se,ATSC2se,ATSC3se,ATSC4se,ATSC5se,ATSC6se,ATSC7se,ATSC8se,ATSC0pe,ATSC1pe,ATSC2pe,ATSC3pe,ATSC4pe,ATSC5pe,ATSC6pe,ATSC7pe,ATSC8pe,ATSC0are,ATSC1are,ATSC2are,ATSC3are,ATSC4are,ATSC5are,ATSC6are,ATSC7are,ATSC8are,ATSC0p,ATSC1p,ATSC2p,ATSC3p,ATSC4p,ATSC5p,ATSC6p,ATSC7p,ATSC8p,ATSC0i,ATSC1i,ATSC2i,ATSC3i,ATSC4i,ATSC5i,ATSC6i,ATSC7i,ATSC8i,AATSC0c,AATSC1c,AATSC2c,AATSC3c,AATSC4c,AATSC5c,AATSC6c,AATSC7c,AATSC8c,AATSC0dv,AATSC1dv,AATSC2dv,AATSC3dv,AATSC4dv,AATSC5dv,AATSC6dv,AATSC7dv,AATSC8dv,AATSC0d,AATSC1d,AATSC2d,AATSC3d,AATSC4d,AATSC5d,AATSC6d,AATSC7d,AATSC8d,AATSC0s,AATSC1s,AATSC2s,AATSC3s,AATSC4s,AATSC5s,AATSC6s,AATSC7s,AATSC8s,AATSC0Z,AATSC1Z,AATSC2Z,AATSC3Z,AATSC4Z,AATSC5Z,AATSC6Z,AATSC7Z,AATSC8Z,AATSC0m,AATSC1m,AATSC2m,AATSC3m,AATSC4m,AATSC5m,AATSC6m,AATSC7m,AATSC8m,AATSC0v,AATSC1v,AATSC2v,AATSC3v,AATSC4v,AATSC5v,AATSC6v,AATSC7v,AATSC8v,AATSC0se,AATSC1se,AATSC2se,AATSC3se,AATSC4se,AATSC5se,AATSC6se,AATSC7se,AATSC8se,AATSC0pe,AATSC1pe,AATSC2pe,AATSC3pe,AATSC4pe,AATSC5pe,AATSC6pe,AATSC7pe,AATSC8pe,AATSC0are,AATSC1are,AATSC2are,AATSC3are,AATSC4are,AATSC5are,AATSC6are,AATSC7are,AATSC8are,AATSC0p,AATSC1p,AATSC2p,AATSC3p,AATSC4p,AATSC5p,AATSC6p,AATSC7p,AATSC8p,AATSC0i,AATSC1i,AATSC2i,AATSC3i,AATSC4i,AATSC5i,AATSC6i,AATSC7i,AATSC8i,MATS1c,MATS2c,MATS3c,MATS4c,MATS5c,MATS6c,MATS7c,MATS8c,MATS1dv,MATS2dv,MATS3dv,MATS4dv,MATS5dv,MATS6dv,MATS7dv,MATS8dv,MATS1d,MATS2d,MATS3d,MATS4d,MATS5d,MATS6d,MATS7d,MATS8d,MATS1s,MATS2s,MATS3s,MATS4s,MATS5s,MATS6s,MATS7s,MATS8s,MATS1Z,MATS2Z,MATS3Z,MATS4Z,MATS5Z,MATS6Z,MATS7Z,MATS8Z,MATS1m,MATS2m,MATS3m,MATS4m,MATS5m,MATS6m,MATS7m,MATS8m,MATS1v,MATS2v,MATS3v,MATS4v,MATS5v,MATS6v,MATS7v,MATS8v,MATS1se,MATS2se,MATS3se,MATS4se,MATS5se,MATS6se,MATS7se,MATS8se,MATS1pe,MATS2pe,MATS3pe,MATS4pe,MATS5pe,MATS6pe,MATS7pe,MATS8pe,MATS1are,MATS2are,MATS3are,MATS4are,MATS5are,MATS6are,MATS7are,MATS8are,MATS1p,MATS2p,MATS3p,MATS4p,MATS5p,MATS6p,MATS7p,MATS8p,MATS1i,MATS2i,MATS3i,MATS4i,MATS5i,MATS6i,MATS7i,MATS8i,GATS1c,GATS2c,GATS3c,GATS4c,GATS5c,GATS6c,GATS7c,GATS8c,GATS1dv,GATS2dv,GATS3dv,GATS4dv,GATS5dv,GATS6dv,GATS7dv,GATS8dv,GATS1d,GATS2d,GATS3d,GATS4d,GATS5d,GATS6d,GATS7d,GATS8d,GATS1s,GATS2s,GATS3s,GATS4s,GATS5s,GATS6s,GATS7s,GATS8s,GATS1Z,GATS2Z,GATS3Z,GATS4Z,GATS5Z,GATS6Z,GATS7Z,GATS8Z,GATS1m,GATS2m,GATS3m,GATS4m,GATS5m,GATS6m,GATS7m,GATS8m,GATS1v,GATS2v,GATS3v,GATS4v,GATS5v,GATS6v,GATS7v,GATS8v,GATS1se,GATS2se,GATS3se,GATS4se,GATS5se,GATS6se,GATS7se,GATS8se,GATS1pe,GATS2pe,GATS3pe,GATS4pe,GATS5pe,GATS6pe,GATS7pe,GATS8pe,GATS1are,GATS2are,GATS3are,GATS4are,GATS5are,GATS6are,GATS7are,GATS8are,GATS1p,GATS2p,GATS3p,GATS4p,GATS5p,GATS6p,GATS7p,GATS8p,GATS1i,GATS2i,GATS3i,GATS4i,GATS5i,GATS6i,GATS7i,GATS8i,BCUTc-1h,BCUTc-1l,BCUTdv-1h,BCUTdv-1l,BCUTd-1h,BCUTd-1l,BCUTs-1h,BCUTs-1l,BCUTZ-1h,BCUTZ-1l,BCUTm-1h,BCUTm-1l,BCUTv-1h,BCUTv-1l,BCUTse-1h,BCUTse-1l,BCUTpe-1h,BCUTpe-1l,BCUTare-1h,BCUTare-1l,BCUTp-1h,BCUTp-1l,BCUTi-1h,BCUTi-1l,BalabanJ,SpAbs_DzZ,SpMax_DzZ,SpDiam_DzZ,SpAD_DzZ,SpMAD_DzZ,LogEE_DzZ,SM1_DzZ,VE1_DzZ,VE2_DzZ,VE3_DzZ,VR1_DzZ,VR2_DzZ,VR3_DzZ,SpAbs_Dzm,SpMax_Dzm,SpDiam_Dzm,SpAD_Dzm,SpMAD_Dzm,LogEE_Dzm,SM1_Dzm,VE1_Dzm,VE2_Dzm,VE3_Dzm,VR1_Dzm,VR2_Dzm,VR3_Dzm,SpAbs_Dzv,SpMax_Dzv,SpDiam_Dzv,SpAD_Dzv,SpMAD_Dzv,LogEE_Dzv,SM1_Dzv,VE1_Dzv,VE2_Dzv,VE3_Dzv,VR1_Dzv,VR2_Dzv,VR3_Dzv,SpAbs_Dzse,SpMax_Dzse,SpDiam_Dzse,SpAD_Dzse,SpMAD_Dzse,LogEE_Dzse,SM1_Dzse,VE1_Dzse,VE2_Dzse,VE3_Dzse,VR1_Dzse,VR2_Dzse,VR3_Dzse,SpAbs_Dzpe,SpMax_Dzpe,SpDiam_Dzpe,SpAD_Dzpe,SpMAD_Dzpe,LogEE_Dzpe,SM1_Dzpe,VE1_Dzpe,VE2_Dzpe,VE3_Dzpe,VR1_Dzpe,VR2_Dzpe,VR3_Dzpe,SpAbs_Dzare,SpMax_Dzare,SpDiam_Dzare,SpAD_Dzare,SpMAD_Dzare,LogEE_Dzare,SM1_Dzare,VE1_Dzare,VE2_Dzare,VE3_Dzare,VR1_Dzare,VR2_Dzare,VR3_Dzare,SpAbs_Dzp,SpMax_Dzp,SpDiam_Dzp,SpAD_Dzp,SpMAD_Dzp,LogEE_Dzp,SM1_Dzp,VE1_Dzp,VE2_Dzp,VE3_Dzp,VR1_Dzp,VR2_Dzp,VR3_Dzp,SpAbs_Dzi,SpMax_Dzi,SpDiam_Dzi,SpAD_Dzi,SpMAD_Dzi,LogEE_Dzi,SM1_Dzi,VE1_Dzi,VE2_Dzi,VE3_Dzi,VR1_Dzi,VR2_Dzi,VR3_Dzi,BertzCT,nBonds,nBondsO,nBondsS,nBondsD,nBondsT,nBondsA,nBondsM,nBondsKS,nBondsKD,RNCG,RPCG,C1SP1,C2SP1,C1SP2,C2SP2,C3SP2,C1SP3,C2SP3,C3SP3,C4SP3,HybRatio,FCSP3,Xch-3d,Xch-4d,Xch-5d,Xch-6d,Xch-7d,Xch-3dv,Xch-4dv,Xch-5dv,Xch-6dv,Xch-7dv,Xc-3d,Xc-4d,Xc-5d,Xc-6d,Xc-3dv,Xc-4dv,Xc-5dv,Xc-6dv,Xpc-4d,Xpc-5d,Xpc-6d,Xpc-4dv,Xpc-5dv,Xpc-6dv,Xp-0d,Xp-1d,Xp-2d,Xp-3d,Xp-4d,Xp-5d,Xp-6d,Xp-7d,AXp-0d,AXp-1d,AXp-2d,AXp-3d,AXp-4d,AXp-5d,AXp-6d,AXp-7d,Xp-0dv,Xp-1dv,Xp-2dv,Xp-3dv,Xp-4dv,Xp-5dv,Xp-6dv,Xp-7dv,AXp-0dv,AXp-1dv,AXp-2dv,AXp-3dv,AXp-4dv,AXp-5dv,AXp-6dv,AXp-7dv,SZ,Sm,Sv,Sse,Spe,Sare,Sp,Si,MZ,Mm,Mv,Mse,Mpe,Mare,Mp,Mi,SpAbs_Dt,SpMax_Dt,SpDiam_Dt,SpAD_Dt,SpMAD_Dt,LogEE_Dt,SM1_Dt,VE1_Dt,VE2_Dt,VE3_Dt,VR1_Dt,VR2_Dt,VR3_Dt,DetourIndex,SpAbs_D,SpMax_D,SpDiam_D,SpAD_D,SpMAD_D,LogEE_D,VE1_D,VE2_D,VE3_D,VR1_D,VR2_D,VR3_D,NsLi,NssBe,NssssBe,NssBH,NsssB,NssssB,NsCH3,NdCH2,NssCH2,NtCH,NdsCH,NaaCH,NsssCH,NddC,NtsC,NdssC,NaasC,NaaaC,NssssC,NsNH3,NsNH2,NssNH2,NdNH,NssNH,NaaNH,NtN,NsssNH,NdsN,NaaN,NsssN,NddsN,NaasN,NssssN,NsOH,NdO,NssO,NaaO,NsF,NsSiH3,NssSiH2,NsssSiH,NssssSi,NsPH2,NssPH,NsssP,NdsssP,NsssssP,NsSH,NdS,NssS,NaaS,NdssS,NddssS,NsCl,NsGeH3,NssGeH2,NsssGeH,NssssGe,NsAsH2,NssAsH,NsssAs,NsssdAs,NsssssAs,NsSeH,NdSe,NssSe,NaaSe,NdssSe,NddssSe,NsBr,NsSnH3,NssSnH2,NsssSnH,NssssSn,NsI,NsPbH3,NssPbH2,NsssPbH,NssssPb,SsLi,SssBe,SssssBe,SssBH,SsssB,SssssB,SsCH3,SdCH2,SssCH2,StCH,SdsCH,SaaCH,SsssCH,SddC,StsC,SdssC,SaasC,SaaaC,SssssC,SsNH3,SsNH2,SssNH2,SdNH,SssNH,SaaNH,StN,SsssNH,SdsN,SaaN,SsssN,SddsN,SaasN,SssssN,SsOH,SdO,SssO,SaaO,SsF,SsSiH3,SssSiH2,SsssSiH,SssssSi,SsPH2,SssPH,SsssP,SdsssP,SsssssP,SsSH,SdS,SssS,SaaS,SdssS,SddssS,SsCl,SsGeH3,SssGeH2,SsssGeH,SssssGe,SsAsH2,SssAsH,SsssAs,SsssdAs,SsssssAs,SsSeH,SdSe,SssSe,SaaSe,SdssSe,SddssSe,SsBr,SsSnH3,SssSnH2,SsssSnH,SssssSn,SsI,SsPbH3,SssPbH2,SsssPbH,SssssPb,MAXsLi,MAXssBe,MAXssssBe,MAXssBH,MAXsssB,MAXssssB,MAXsCH3,MAXdCH2,MAXssCH2,MAXtCH,MAXdsCH,MAXaaCH,MAXsssCH,MAXddC,MAXtsC,MAXdssC,MAXaasC,MAXaaaC,MAXssssC,MAXsNH3,MAXsNH2,MAXssNH2,MAXdNH,MAXssNH,MAXaaNH,MAXtN,MAXsssNH,MAXdsN,MAXaaN,MAXsssN,MAXddsN,MAXaasN,MAXssssN,MAXsOH,MAXdO,MAXssO,MAXaaO,MAXsF,MAXsSiH3,MAXssSiH2,MAXsssSiH,MAXssssSi,MAXsPH2,MAXssPH,MAXsssP,MAXdsssP,MAXsssssP,MAXsSH,MAXdS,MAXssS,MAXaaS,MAXdssS,MAXddssS,MAXsCl,MAXsGeH3,MAXssGeH2,MAXsssGeH,MAXssssGe,MAXsAsH2,MAXssAsH,MAXsssAs,MAXsssdAs,MAXsssssAs,MAXsSeH,MAXdSe,MAXssSe,MAXaaSe,MAXdssSe,MAXddssSe,MAXsBr,MAXsSnH3,MAXssSnH2,MAXsssSnH,MAXssssSn,MAXsI,MAXsPbH3,MAXssPbH2,MAXsssPbH,MAXssssPb,MINsLi,MINssBe,MINssssBe,MINssBH,MINsssB,MINssssB,MINsCH3,MINdCH2,MINssCH2,MINtCH,MINdsCH,MINaaCH,MINsssCH,MINddC,MINtsC,MINdssC,MINaasC,MINaaaC,MINssssC,MINsNH3,MINsNH2,MINssNH2,MINdNH,MINssNH,MINaaNH,MINtN,MINsssNH,MINdsN,MINaaN,MINsssN,MINddsN,MINaasN,MINssssN,MINsOH,MINdO,MINssO,MINaaO,MINsF,MINsSiH3,MINssSiH2,MINsssSiH,MINssssSi,MINsPH2,MINssPH,MINsssP,MINdsssP,MINsssssP,MINsSH,MINdS,MINssS,MINaaS,MINdssS,MINddssS,MINsCl,MINsGeH3,MINssGeH2,MINsssGeH,MINssssGe,MINsAsH2,MINssAsH,MINsssAs,MINsssdAs,MINsssssAs,MINsSeH,MINdSe,MINssSe,MINaaSe,MINdssSe,MINddssSe,MINsBr,MINsSnH3,MINssSnH2,MINsssSnH,MINssssSn,MINsI,MINsPbH3,MINssPbH2,MINsssPbH,MINssssPb,ECIndex,ETA_alpha,AETA_alpha,ETA_shape_p,ETA_shape_y,ETA_shape_x,ETA_beta,AETA_beta,ETA_beta_s,AETA_beta_s,ETA_beta_ns,AETA_beta_ns,ETA_beta_ns_d,AETA_beta_ns_d,ETA_eta,AETA_eta,ETA_eta_L,AETA_eta_L,ETA_eta_R,AETA_eta_R,ETA_eta_RL,AETA_eta_RL,ETA_eta_F,AETA_eta_F,ETA_eta_FL,AETA_eta_FL,ETA_eta_B,AETA_eta_B,ETA_eta_BR,AETA_eta_BR,ETA_dAlpha_A,ETA_dAlpha_B,ETA_epsilon_1,ETA_epsilon_2,ETA_epsilon_3,ETA_epsilon_4,ETA_epsilon_5,ETA_dEpsilon_A,ETA_dEpsilon_B,ETA_dEpsilon_C,ETA_dEpsilon_D,ETA_dBeta,AETA_dBeta,ETA_psi_1,ETA_dPsi_A,ETA_dPsi_B,fragCpx,fMF,nHBAcc,nHBDon,IC0,IC1,IC2,IC3,IC4,IC5,TIC0,TIC1,TIC2,TIC3,TIC4,TIC5,SIC0,SIC1,SIC2,SIC3,SIC4,SIC5,BIC0,BIC1,BIC2,BIC3,BIC4,BIC5,CIC0,CIC1,CIC2,CIC3,CIC4,CIC5,MIC0,MIC1,MIC2,MIC3,MIC4,MIC5,ZMIC0,ZMIC1,ZMIC2,ZMIC3,ZMIC4,ZMIC5,Kier1,Kier2,Kier3,FilterItLogS,VMcGowan,LabuteASA,PEOE_VSA1,PEOE_VSA2,PEOE_VSA3,PEOE_VSA4,PEOE_VSA5,PEOE_VSA6,PEOE_VSA7,PEOE_VSA8,PEOE_VSA9,PEOE_VSA10,PEOE_VSA11,PEOE_VSA12,PEOE_VSA13,SMR_VSA1,SMR_VSA2,SMR_VSA3,SMR_VSA4,SMR_VSA5,SMR_VSA6,SMR_VSA7,SMR_VSA8,SMR_VSA9,SlogP_VSA1,SlogP_VSA2,SlogP_VSA3,SlogP_VSA4,SlogP_VSA5,SlogP_VSA6,SlogP_VSA7,SlogP_VSA8,SlogP_VSA9,SlogP_VSA10,SlogP_VSA11,EState_VSA1,EState_VSA2,EState_VSA3,EState_VSA4,EState_VSA5,EState_VSA6,EState_VSA7,EState_VSA8,EState_VSA9,EState_VSA10,VSA_EState1,VSA_EState2,VSA_EState3,VSA_EState4,VSA_EState5,VSA_EState6,VSA_EState7,VSA_EState8,VSA_EState9,MDEC-11,MDEC-12,MDEC-13,MDEC-14,MDEC-22,MDEC-23,MDEC-24,MDEC-33,MDEC-34,MDEC-44,MDEO-11,MDEO-12,MDEO-22,MDEN-11,MDEN-12,MDEN-13,MDEN-22,MDEN-23,MDEN-33,MID,AMID,MID_h,AMID_h,MID_C,AMID_C,MID_N,AMID_N,MID_O,AMID_O,MID_X,AMID_X,MPC2,MPC3,MPC4,MPC5,MPC6,MPC7,MPC8,MPC9,MPC10,TMPC10,piPC1,piPC2,piPC3,piPC4,piPC5,piPC6,piPC7,piPC8,piPC9,piPC10,TpiPC10,apol,bpol,nRing,n3Ring,n4Ring,n5Ring,n6Ring,n7Ring,n8Ring,n9Ring,n10Ring,n11Ring,n12Ring,nG12Ring,nHRing,n3HRing,n4HRing,n5HRing,n6HRing,n7HRing,n8HRing,n9HRing,n10HRing,n11HRing,n12HRing,nG12HRing,naRing,n3aRing,n4aRing,n5aRing,n6aRing,n7aRing,n8aRing,n9aRing,n10aRing,n11aRing,n12aRing,nG12aRing,naHRing,n3aHRing,n4aHRing,n5aHRing,n6aHRing,n7aHRing,n8aHRing,n9aHRing,n10aHRing,n11aHRing,n12aHRing,nG12aHRing,nARing,n3ARing,n4ARing,n5ARing,n6ARing,n7ARing,n8ARing,n9ARing,n10ARing,n11ARing,n12ARing,nG12ARing,nAHRing,n3AHRing,n4AHRing,n5AHRing,n6AHRing,n7AHRing,n8AHRing,n9AHRing,n10AHRing,n11AHRing,n12AHRing,nG12AHRing,nFRing,n4FRing,n5FRing,n6FRing,n7FRing,n8FRing,n9FRing,n10FRing,n11FRing,n12FRing,nG12FRing,nFHRing,n4FHRing,n5FHRing,n6FHRing,n7FHRing,n8FHRing,n9FHRing,n10FHRing,n11FHRing,n12FHRing,nG12FHRing,nFaRing,n4FaRing,n5FaRing,n6FaRing,n7FaRing,n8FaRing,n9FaRing,n10FaRing,n11FaRing,n12FaRing,nG12FaRing,nFaHRing,n4FaHRing,n5FaHRing,n6FaHRing,n7FaHRing,n8FaHRing,n9FaHRing,n10FaHRing,n11FaHRing,n12FaHRing,nG12FaHRing,nFARing,n4FARing,n5FARing,n6FARing,n7FARing,n8FARing,n9FARing,n10FARing,n11FARing,n12FARing,nG12FARing,nFAHRing,n4FAHRing,n5FAHRing,n6FAHRing,n7FAHRing,n8FAHRing,n9FAHRing,n10FAHRing,n11FAHRing,n12FAHRing,nG12FAHRing,nRot,RotRatio,SLogP,SMR,TopoPSA(NO),TopoPSA,GGI1,GGI2,GGI3,GGI4,GGI5,GGI6,GGI7,GGI8,GGI9,GGI10,JGI1,JGI2,JGI3,JGI4,JGI5,JGI6,JGI7,JGI8,JGI9,JGI10,JGT10,Diameter,Radius,TopoShapeIndex,PetitjeanIndex,Vabc,VAdjMat,MWC01,MWC02,MWC03,MWC04,MWC05,MWC06,MWC07,MWC08,MWC09,MWC10,TMWC10,SRW02,SRW03,SRW04,SRW05,SRW06,SRW07,SRW08,SRW09,SRW10,TSRW10,MW,AMW,WPath,WPol,Zagreb1,Zagreb2,mZagreb1,mZagreb2 \clearpage \bibliography{bibliography-prescreening} \end{document}