[ last-release ]

Version 3.9-4a

• Added Label filter expression in Plot (`make-plot) to control label visibility
• Added Kennard Stone Train/Test Split method (Tools/Table)
• fixed crash in find database command
• fixed search by indexed fields: spfas.ID == "NMNA3_HUMAN"
• Added support for large object stacks (32 bit integer overflow)
• some speed optimization in MCS
• Fixed crash in GINGER for some rare cases
• Improvements in GINGER: better macro cycle and fused ring treatment
• fixes in CHEMFILTER.sdf (chemical alerts)
• New GPU accelerated formal charge assignment using NN pKa models and protonation state system equilibrium
• Added new NN models for pKa predictions ( MODELS.MolpKaAcid = "pKaEnsAcid"; MODELS.MolpKaBase = "pKaEnsBase" )
• fixes in editing reaction with recursive SMARTS in GUI
• fixed auto bond order assignment for unknown residues in the chain
• fixes in read pdb cif format
• fixed in reading some NMR pdb
• added checkpoint= to _dockScan script to automatically restore from last interrupted position
• fixes in Pi-Pi 2D interaction depiction
• fixed bug in read pdb which caused hanging for certain entries (8p0t)
• added "checkpoint=" argument to _dockScan to automatically start from last position if job was killed and then restarted
• added "convRing" neighbour aggregation mode to perform 1-D convolution on macro cycles
• Converted Wuxi reactions to Molsoft Space format (fast MEL enumeration, fast diversity subset enumeration, CombiRidge, etc...)
• added heavy option to learn command to dump descriptors and graph to file and efficiently use it in learning.
• added ability to rescore top N conformations by CombiScore (VLS+RTCNN) in RIDGE (confsRescore= argument)
• Fixed in linker search in conformational DB (added checking of valency rules)
• Fixed object range selection ( a_$i1:$i2 ) for large number of objects (>32K).
• added Descriptor( pharmacophore ) function to generate fingerprint based in hashed triplets of ph4 features

Version 3.9-4

• GINGER: Graph Internal-coordinate Neural-network conformer Generator with Energy Refinement for Large Chemical Libraries
  • state-of-the-art generative Neural Network-based algorithm to efficiently explore the conformational space of molecules, providing diverse, low-energy conformers
  • Generate conformers for a 10 mln compound library in one day on a single RTX4090 'gamer' workstation
  • Produces highly compressed conformational database which can be used as an input for RIDE and RIDGE
  • Check related publication describing the method
• RIDGE: Rapid Docking GPU Engine
  • Extremely fast (GPU accelarated) and accurate structure-based virtual ligand screening method.
  • ~100 compounds/sec docking speed (RTX 4090)
  • Comparable (< ~10% loss in docking success rate) quality of the poses with flexible ICM for mid-sized ligands
• GigaScreen Deep Learning Approach for Screening Ultra Large Databases
  • The GigaScreen method combines machine learning and deep learning tools to tackle the computational intensity of screening very large chemical databases.
  • One of the advantages of the GigaScreen protocol is that large giga-sized libraries can be screened on a single box, eliminating the need for expensive computational resources or additional software.
• New features
  • added Profile function to calculate Statistical Torsion Profiles from chemical structure
  • experimental Max ARM64 native package (OpenMM MD support)
  • added OpenCL (GPU) support for OpenMM on Windows, Linux and Mac (arm64)
  • added e3dAddToolsmacto to add "vls-hitlist-style" panel to Ligand Editor tables
  • added legend to 2D Ligand interaction diagrams
  • added plot color by point density
  • added nProc support to make flat command
  • added scanConvertTable2Hitlist macro to convert "external" table with poses to ICM hitlist
  • added -stride argument to _dockScan (from/stride chunking)
  • added superimpose M_xyz1 M_xyz2 and Transform( M_xyz R_vec12 )
  • new GIGA search databases: REAL 2022 (5.16B) and ZINC20 (1.79B)
  • added File.Load.AlphaFold Structure
  • added support for "mixed" enumeration of linkers and R-groups in enumerate library command
  • added 'Static' option to "Search Pharmacophore" dialog
  • added _ph4 script for pharmacophore search from the command line (supports compressed RIDE db)
  • added text filter to 'Chemistry/Properties' dialog
  • added per atom RTCNN contribution (`scoreEval_nnInterMod macro)
  • added support of multiple GPU for OpenMM MD simulations (GUI and command line)
  • added optional Enumeration/Prediction step to Free Wilson Analysis
  • added atom option for predict command to return per-atom contributions from NN models
  • added rename-peptide-autocommand
  • added Index( X center ) to efficiently pick diverse subset using MinMax algorithm
  • added Max( sarray ) to return longest common prefix of sarray elements
• Fixes
  • fixed some rare stereo chemistry inversion in some bicycles.
  • improvements in alignment ruler rendering
  • fixes and improvements in pi-pi and pi-cation interactions list (Added label support)
  • fixes in SDF V30 reader
  • fixes for Mac OSX Sonoma
  • speed improvement in convert for small molecules
  • fixed crash in on Mac when using MolEdit in modal mode
  • added proper parsing of PDB 'COMPND' field
  • added auto rename of PDB molecules starting from digit.
  • fixes in read sarray can read from index ( read sarray uniprot_sprot.ID == "IL2_HUMAN" )
  • plot point labels now respect column format (e.g: number of significant figures)
  • some crash fixes in alignment view
  • improvements in alignment profile visualisation and residue number labeling
  • fixed incorrect atom mapping find find molecule sstructure
  • "Docking/Make Hit List" will import original columns by default
  • fixes in 'Hetero Scan' option in Ligand Editor (crash + menu disabled)
  • Plot: continuous rainbow widget in legend (for numerical column coloring)
  • fixed crash in IUPAC
  • fixed calcBBBScore function to ignore salts
  • optimized clustering for large data sets
  • fixed bug in swiss index makeing and entry access
  • added more thorough hydrogen orientation optimization in LE
  • fixed crash in converting torsion list to table
  • convertObject preserves stack with 'Replace the Original' option
  • fixes in importExcel macro
  • fixes in multiple GL window support
  • fixes and improvements in MCS algorithm
  • fixes in indexing problem of large SDF files on Windows
  • fixes in "Chemistry/Calculate Properties/Most Basic:Acidic" calculation for certain molecules
  • fixed bug in using fork in for loops
  • 'Docking/Dock Chemical Table' joins original columns by default
  • Linker search now works correctly with covalent setup in Ligand Editor
  • Some improvements in Linker search with 3 cutting points
  • TOOLS.maxHeavyAtomsForDepict2D is used in 2D editing in LigEdit
  • fixed bug in "Open in LigEdit" button in Docking Hit list
  • fixed empty docking log with nproc>1 on Linux/Mac
  • fixed rare crash in show energy command (when "el" term is not set)
  • improvement in coloring chemicals by model
  • added several fused ring depiction templates
  • original molecule name is shown in the comment in VLS hitlist

Version 3.9-3a

• RIDE support multiple DB input from both GUI and command line
• parity stereo is correctly interpreted from MOL format with 0D coordinates
• introduced compressed more efficient RIDE DB format (4-6x times less disk space, much faster reading speed)
• added Split( residue ) function
• removed length limitation in label field in molecular objects (`Label{Label()})
• Fixed auto detection of numerical column types in Excel import macro
• Fixed problem in Linker and Group can with non-unique output file name for background jobs
• LMX licensing support (https://www.x-formation.com/)
• Markush enumeration (`enumerate-library) now supports multiple positions for the same R-group
• Fixed few memory issues in Ligedit Find R-group
• Added GUI for Tools.3D Similarity PDB search
• added ECFP fingerprint support for SALI analysis
• added support for .cxsmiles format to GUI (File/Open)
• added command to split grob by plane (see split grob)
• some fixes in chemCalcPKA macro
• fixes in maximum common substructure algorithm
• switched to new Uniprot REST API
• add multi-thread support (`nProc) for most of built-in functions in add column function expressions (GUI: Chemical/Calculate Properties)
• add MolCalcContacts KNIME node to calculate interactions lists for a set of 3D ligands to a receptor
• optimization GUI table for large tables
• fixed problem with reading large files
• fixed "truncation" when importing large numbers from Excel
• added support for embedded stack for chemical arrays (see concat_conf aggregation for group column command and split commands)
• fixed bug in find molecule introduced in earlier builds of 392e
• updates in historeceptomicsAPI
• Added support to define stereo for attachment point
• Fixed bug in RIDE GUI dialog. (multiple template selection was ignored and only first molecule used in search)
• find table command supports nProc
• added Xyz( as ring [all]) to return ring centers (with all option will return flexible rings too, otherwise flat only)
• added Grob( M_xyz, r_ra, ball )
• Automatic update of ChEMBL index
• Improvements in "Protonation States vs pH" functionality (isoelectric point, pH vs charge graph, etc ..)
• Fixed incorrect zero occupancy treatment in RIDE search
• Fixed incorrect charge setting for guanidine when converting 3D -> 2D
• background job with multiple processes now return full output from all jobs
• added "R/S" to atomLabelStyle property to display stereo R/S labels
• Fixed truncation of long numeric properties (e.g: ID numbers)
• Added mol2img.icm script to export chemical(s) to image(s)
• Fixed (In docking/covalent - if I try and delete a reaction or edit ICM hangs)
• MolScreen moved to top level menu
• added hint how to connect ring as sprio in moledit
• added "Sort By Frequency" option for Histogram by string columns
• added Peptide view for table column with icm sequences (modified side chains, staples, disulfide bonds, etc ...)
• fixed some bugs in 2D ligand interactions diagrams
• fixed certain issues in plotting protonation states vs pH
• cache for amino compressed view (improved speed)
• added notarization to Mac package
• fixed axis label order in plot
• fixed atom coloring by model (`color-chemical)
• new Enamine REAL release 2021q12
• rename peptide command
• fixed bug with residue order in assign residue command
• added initial Molecular Dynamics support using OpenMM package (`dynamics command)
• API and GUI to Historeceptomics server
• added TOOLS TOOLS.mcsThoroughness parameter for tuning MCS (Maximum Common Substructure) functions
• add ability to layout tabs in extra table panels from script. ('toolsLayout' header)
• fixed converting to 3D compounds with chiral centers certain ring systems
• fixed _confGen for Boron containing compounds
• updated COD to the latest version
• optimized speed and memory usage for multiple ligand template in RIDE
• fixed plot axis labels overlap issue
• fixed some issues with Excel export
• fixed problem with running RIDE on multiple GPUs
• fix large margins for pie inline plots in table
• added 'front' property for 2D chemical bond to emulate "depth" in 2D
• fixed binary read/write for atom colors in chemical table
• added 'Rigid' option to GUI APF Superposition to Template
• fixes and speed improvements in torsion profile API (optimized _confGen)
• Fixed some problems with pi-pi stack display
• BIOMT, CRYST1, SCALE, MTRIX, ORIGX support for PDB mmCIF
• added -shape option to _chemSuper to match shape only
• fixed bug in _chemSuper (Xweight wasn't treated with -e option)

Version 3.9-2d

• New Torsion Profile ANN model.
• fixed some issues in nested for loops with fork option
• fixed some issues when running RIDE on Windows using large DBs
• fixed some font issues in 3D on Mac
• Added Select( molecule ) select i-th disconnected fragment
• Added value= label for plots with matrix (on mouse over)
• Added Chemistry.Combinatorial Chemistry.Tools.Convert Attachment Points to R-groups and Convert R-groups to Attachment Points menu items
• Fixed some font size issues on high resolution monitors
• Fixed a bug docking with "Charge groups" = "sample" option is on
• Added "Advanced/Evaluate Global Strain" for LIGANDS table in LE
• highlight icm torsion and display profile when rotating
• added 'S[123]' SMARTS attribute to match absolute S=1 R=2 stereo. Nof( t.mol "[C;S1;$(C(N)C=O)]") L-amino
• fixed label display issue on Mac in 3D
• Transpose table function preserves color and format and fully reversible
• Added shortcuts to ECFP3,ECFP4,ECFP5,ECFP6 in learn dialog
• fixed several visualisation issues on Mac ( checkbox size, row labels, popup sub-menu width, non-modal dialog stays on top )
• ability to draw chemicals in legend view
• improved default column width for columns with spaces
• added SMIRKS support for reaction mechanism
• fixed "blank" workspace while running icm command
• added 'model' preference to GRAPHICS.varRingColorStyle
• added predict option to set variable command to assign torsion values by model
• fixed sstructure option in display stack command
• some fixed in Mac version: tree open sign is now visible, disabled apple full screen mode because it causes some issues with Qt version.
• Improvements in "Multiple position R-group scan" (background option, ability to provide custom R-group table)
• Fixed "Application Unable to Start Correctly Error (0xc000007b)" error on Windows on certain system configurations
• Added "Exit points" to LE
• improvements in handling error 8808 (some atom coordinates are NAN)
• updated MolPSA and MolVolume models (PLS, ECFP4, Q2=0.99)
• fixed crash in set vrestraint with only option introduced in 3.9-2c
• add Chemical( X chiral inverse ) to invert chirality for both 2D and 3D molecules
• surface accessible area view in LE
• truncation indicator for bars in plots
• added cis/trans enforcement in the reaction product
• added support of "distance" bond with constraints in reaction
• fixed bug in histogram in "split by column" mode
• fixed copy/paste column issues in spreadsheets
• fixed some issues in RIDE GPU code
• fixed online dictionary search behavior in 2D editor for certain cases
• fixed some bugs in PostgreSQL client
• 'Linker Database Scan' works on disconnected fragments

Version 3.9-2c

• fixed label positioning in pie charts
• fixed incorrect "cis/trans" depiction of wedged bonds around ring for certain cases.
• fixed inability to clear hydrogen bond label after once displayed (GRAPHICS.hbondStyle preference)
• fixed crash in learn command
• Unified R-group Scan and Linker Scan dialog in LE (with parallel option)
• Fragment docking in LE to fill pocket with fragments
• Fixed some visual glitches in ICM for Mac on Big Sur: fonts in popup menus, invisible text in active tab
• added 'Bring To Front' menu for table popup. Useful to navigate through large number of tables in workspace
• fixed redirection for large PDB entries to CIF
• fixed several issues with editing 3D compounds in chemical table.
• fixed some issues in reactive cysteine identification (macro reactiveCys)
• added ability to annotate spreadsheets by adding Legend (new tab in tools panel)
• fixes to support RIDE search in ultra large DBs
• added new option for GRAPHICS.varRingColorStylepreference to color torsion by statistics from TorLib.xml library
• added Mass( | [exact] ) function to calculate chemical weight directly from formula;
• added Info( filename ) function to return collection with MOLT file properties (E.g: supports substructure search, RIDE, etc ..)
• updated MolLogS model: R2=0.97 Q2=0.83 (OOB cross validation), ECFP4 binary+MolLogP, Random Forest Regression, 9923 compounds training set
• fixed memory allocation in Distance function for large matrices
• compare-molcartcommand for efficient chemical database comparison by exact match or similarity
• importADGridMap macro to import and contour fragment maps (SILCSBio)
• unified linker and R-group substitution
• improved predict option to rank tautomers using pKa model from enumerate tautomer command
• Generate protonation states at given ph4 and concentration percentage cutoff (`protonation-states-table)
• Plot Protonation States vs pH. Right click on molecule, menu Chemistry/Plot Protonation States vs pH. (`protonation-states-table)
• shortcut for small hetero cycle and fused system in group scan ligedit
• description for tables (goes into tooltip help on mouse over)
• fixed duplicated values for ID column when inserting rows in the middle of the table
• Correct current directory is set on MacOS BigSur
• Fixed crash in LigEdit docking with flexible receptor groups
• added script to selected diverse subset from SDF,SMI,CSV,TSV.
• Changed DrugBank to DrugCentral (no commercial limitations)
• added Qa/Qd interpretation in SMARTS matching for acceptor/donor
• Fixed smiles with cis/trans for parsing and generation for certain cases
• basic CDXML format import support
• RIDE docker container update: latest icm binary, latest icmjs, ability to mount host directory with RIDE DBs (/opt/conf)
• added distance=M_dmt argument to find pharmacophore command to match by intra ph4 distances.
• Added pharmacophore search for RIDE DB format. (Chemistry.Superposition.Pharmacophore Search/Superposition)
• Fixed pharmacophore search bug with multiple optional ph4 points
• Fixed MolMechanics/Mutation with unnatural amino acids
• Fixed crash in make tree with explicit distance argument.
• Explicit option to sync between table in 3D browsing mode and graphics

Version 3.9-2b

• Fixed 'modalSession has been exited prematurely - check for a reentrant call to endModalSession' message on Mac which may lead to crash in certain cases.
• Fixed error in RTCNN model when metal is present in the pocket
• Fixed bug with dotted surface representation in slides.

Version 3.9-2a

• nproc support for docking peptide mode on Windows
• added 'Edit/Add Sequence' menu for 2D chemical editor.
• pi-pi stacking visualisation tool
• better error diagnostic and handling in fork command
• GUI to edit Compressed Chemical View dictionary (Tools menu)
• add support for PDB mmCIF format (http://mmcif.wwpdb.org/pdbx-mmcif-home-page.html)
• NEW Rgroup DB scan dialog with multiple filters
• Some fixes in 3D Ligand Diagrams
• Fixed plot coloring by columns with ND
• NEW Graph convolutional NN model for binding score prediction
• switched to new compiler on Linux and Windows
• SALI analysis supports two tables
• Multiple fixes in chemical alignment by scaffold
• New REAL DB 2020q12
• Added interface to Torsion Library (https://pubs.acs.org/doi/10.1021/acs.jcim.5b00522). make torsion profile [table]
• Plots supports values with < and >
• Updated service for FAST structure lookup by Pubchem CID/SID or CID lookup by structure (Insert Column/Chemistry/Pubchem)
• Ability to define custom step function using 'CustomStepFunc' column in MPO
• Fixed visualisation of MPO step functions with open left or right intervals
• Added macro chemImportSilcsBioPh4 to imported pharmacophore file created by SilcsBio.
• Fixed incorrect interpretation of up/down bond in MOL format V300
• ECFP fingerprints ignore hydrogen atoms
• Fixed crash in plot when table is empty
• Import from Excel supports ChemDraw embedded chemical objects
• Slide description can be shown as HTML
• added chemical 'synthesizability' difficulty model 'MolSynth' based on statistics of ECFP fingerprints from vendor libraries
• pcktmScan.icm script can work with molt DB of pockets (much smaller memory requirement)
• Fixed rare crash in PLS learn command
• Corr( I_ I_ binary ) => {accuracy,precision,recall,F1}
• KNIME node for RIDE script
• added LogD predictor (Predict(t.mol,"MolLogD") and Chemistry/Calculate Properties)
• improved calculation of the pKa of most acidic and most basic groups

Version 3.9-1b

• New features
• Column coloring by unique values for string and integer columns (Column Format Dialog)
• Ligand setup in the ligand editor automatically preserves covalent bonding
• added 'h' property for SMARTS to match number of implicit hydrogen.
• Added "compressed column view" for spreadsheets
• Added Chemistry/Convert/IUPAC to 2D (uses external web-service)
• make reaction can invert stereo centers: CC(=C)N[C@](C(O)=O)([R1])[R2]>>CC(=C)N(C)[C@@](C(O)=O)([R1])[R2]
• R-group decomposition supports linker using distance bonds: c1ccc(cc1)[R1]~~[R2]C(O)=O
• Added Free Wilson Regression SAR analysis. (Chemistry/SAR Analysis/Free Wilson Regression and chemFreeWilsonAnalysis macro)
• Added radar style plot for both inline and main table plots
• Added mkunique_sdf.icm script which creates non-redundant sub-set from multiple SDF and CSV files.
• New MolLogP model: R2=0.98 Q2=0.95 (5-fold cross validation), ECFP4 counted, PLS-regression, 13228 compounds training set
• New MolLogS model: R2=0.97 Q2=0.83 (OOB cross validation), ECFP4 binary+MolLogP, Random Forest Regression, 9923 compounds training set from this publication.
• New online Enamine REAL DB (1.2B compounds)
• added hydrogen option for find molecule sstructure command to find correspondence for hydrogen atoms.
• New core replacement tool in the Ligand Editor (Advanced/Find Linker) Works on the same conformational DB as RIDE search.
• Added Chemical( R_react 'product'|'reactant' ) function to extract components. (Also right click on reaction columns and choose 'Split into components')
• added support for metal coordination bonds
• Added ability to set order for string and chemical labels ('xLabelOrder' and 'yLabelOrder')
• Added plot line thickness control. (lineWidth= and GUI interface)
• New fast rigid APF superposition method with GPU support: RIDE (Rapid Isostere Discovery Engine).
  • _ride script for command line use
  • GUI interface: Chemistry/APF Tools/RIDE Search
  • Docker image (www.docker.com) for RIDE server. (http://molsoft.com/distrib/docker/)
• added correction for an ambiguous drawing of some bicyclics compounds with stereo centers.
• added 'Pubchem' sub-folder in "Insert Column Dialog/Chemical"
• Added D-amino acids support in assign residue command
• Added new BBB score function calcBBBScore (DOI: 10.1021/acs.jmedchem.9b01220)
• IcmJS can be executed in WASM (WebAssembly mode) which gives faster load than ASM.JS
• Bug fixes
• Fixed crash in _chemAlign script
• Fixed rare crash in ligand editor when charging an atom
• Fixed SSL error for certain websites (Linux static build)
• Fixed crash when displaying big cluster tree in GUI
• Fixed opening files on Mac by double-clicking in the Finder.
• Fixed bug in make reaction with stereo option.
• Fixed problem with some shapes in plots on Windows.
• Fixed problem with mouse cursor change in 3D ligand editor on Mac
• Fixed issues with build tautomer command when compounds contains peptide bond in unfavorable state: *C(OH)=N
• Fixed excessive memory requirements in PLS are associated with high PLS cross-validation fold number
• Fixed bug in make reaction to proper handle [R1;A] (R1 matches aliphatic only)
• Fixed bug in calculating distance using Maximum Common Substructure metric Distance( chemA chemB sstructure )
• Fixed covalent setup in ligand editor if receptor contains masked atoms
• Fixed bug in 2D editor when called from 3D ligand editor

Version 3.8-7c

• New features
• New version of Enamine REAL 2019 (718M compounds) is available from the Chemical Search Space
• proc= support for _chemSuper and _dockScan on windows
• Inline plots can be added to Grid View in the spreadsheets
• Column filters expressions support comparison between columns (A<B)
• New GUI for column selection in clustering Heatmap
• Added GaussFit function
• PostreSQL support in Molcart (both server and client side)
• Added hydrogen bond distance label in 2D ligand diagram.
• torScan macro now supports one or two torsions in the input selection
• Added date type support when exporting from Excel
• Added Mean/Median/Interquartile view for plots
• Added 'Log(frequency)' option for histogram plot
• Fixed 'Regression Line' in plot when both axises are logarithmic
• Added Median function
• Compressed Hierarchical uses randomSeed which can be fixed to produce identical results on each run.
• added support for Spotfire STDF format.
• SMILES column can be views as a main column in grid view.
• MPO works on SMILES column
• added support for atom alias field in SDF
• added automatic bond typing for non-standard residues in amino and nuclear chains
• Bug fixes
• Fixed multiple issues int "Export to IcmJS"
• Fixed some errors in "Chemistry/Calculate pKa" (chemCalcPKA macro)
• Multiple fixes for high resolution displays.
• fixed R/S priorities for oxygen double bonded to S or P in tetrahedral configuration.
• Fixed bug in load frame from large GROMACS files
• fixed missing aromatic flag in set tautomer for histidine
• SDF import/export supports long column names

Version 3.8-7b

• fixed incorrect mouse behavior in plot when point labels are displayed
• fixed memory leak in _dockScan
• fixed bug selecting table column from icm.gui when columns have long names (e.g: when comes from Excel)
• Plot options dialog supports column long names
• Improvements in 2D depictions (cis/trans in rings)
• Fixed crash when selecting 'Optimize Ligand Tautomers' option in Convert Object dialog.
• Fixed crash in convert when the N-term is proline with NH3
• added MPO model for Quantitative Estimate of Druglikeness (QED) (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3524573)
• error in child process after fork is properly handled by parent.
• custom menu for maps
• added LIGAND.chargeAfterModifyGroup preference for auto charging ligand after group replacement and group scan
• fixed bioisostere scan in ligand editor for single hydrogen replacement.

Version 3.8-7a

• added TOOLS.pdbChargeNterm option to automatically charge N-terminal residue if it matches the first residue in construct sequence (_SEQRES_ field)
• added option reverse for display stack command
• mkNMstack macro supports metals + some other fixes in normal modes
• convert object correctly handles terminal NH2 and NH3+
• make reaction supports nProc
• Added aromatic contact column in icmPocketFinder which contains percentage of the area exposed to aromatic rings
• Added TLS v1.2 support for https connections
• large file (>2.2GB) support for ICB (64 bit directory)
• fixed bug in memory allocation of large matrices in buffer (e.g: when return from function)
• Improvements in pKa model
• atom-mapping number field is translated to R-groups when reading reactionfrom RXN
• Improvements in Enamine REAL DB online search. (Added interface from Chemical Search space)
• Fixed online drugbank search
• Fixed some issues with make reaction when product contains stereo centers.
• Fixed wrong atom priority assignment for R/S detection in rare situations.
• SALI analysis doesn't require N^2 memory and can work on very large files.
• added 'Ligand Efficiency' calculation into 'Insert Column Dialog'
• fixed bug in conversion of chiral compounds from smiles with coordinates.
• added Srmsd( X_3D_chem ), Srmsd( X_3D_chem1, X_3D_chem2 ), Rmsd( X_3D_chem ) and Rmsd( X_3D_chem1, X_3D_chem2 ) for cross RMSD matrices.
• Torsion customization ( 'totc' in mmff.tot ) now finds overlapped fragments.
• Random Forest classifier supports multiple classes now. (97% accuracy on MNIST digit recognition test http://yann.lecun.com/exdb/mnist/)
• Improvements in 2D depiction.
• Fixed Score( X, X, field ) which return individual APF field components.
• "Chemistry/Select Duplicates" allows to select by non-chemical columns.

Version 3.8-7

• New functionality/Improvements
• Standard way to install/update/remove ICM extensions. (File/Extensions)
• added topological descriptors into Chemistry/Calculate Properties (http://molsoft.com/icm/icm-functions.html#Descriptor-numeric)
• improved torsion rotation tool (changes cursor and shows angle on mouse-over)
• Improved datesupport (GUI to convert, option to parse dates during import from SDF or CSV)
• Added bioisostere replacement in Ligand Editor interface (Advanced/Find Bioisostere)
• Updated bioisostere database
• Self Organizing Networks for chemical space visualisation
• added options to ignore chirality and charges in joining by structure ('Edit/Table/Add External Columns' Dialog)
• Added 'ECFP' option for fingerprints in clustering and PCA analysis.
• Improvements in learn atom for pKa prediction
• Added 'Tautomer Search' option into Chemical Substructure Search. (See also: find table find molcart )
• New modern HTML template (no frames) for exporting to IcmJS html page.
• Add ability to color plot using aggregate functions
• read pdb automatically switches to MMTF (`read-pdb-mmtf) if the entry is not found in PDB format (usually large entries)
• Added support for semi-chiral centers around flexible rings
• Improvements in linker substitution in the ligand editor.
• ability to specify low/high confidence intervals in Plot Options Dialog.
• Added new options to 'Link Column Dialog' ('select matched' and 'show only matched')
• New fonts on Linux in 3D (supports arbitrary sizes and anti-aliasing)
• New functionality scripting
• String( tree ) uses an exiting tree attached to the alignment.
• added operation ( t.bit_column ~ bit_pattern )
• added Nof( "maxFusedAromRings" )
• Added Collection( chemical )
• added charge option for build tautomer command to add charge state enumeration
• added Index( as_ tautomer ) function to return current tautomer state. Takes two arguments: x : current count, y : chain length. E.g: function = "x**(0.5-0.5/y)"
• added 'concat' aggregation function for group column command. Values with the same key will be stored in the array in each cell. (`split-column is opposite operation)
• added new option for covalent docking project l_covMultiple which allows to try multiple covalent mechanisms for single compound.
• Added function=| argument to learn and learn atom commands. Can be applied to counted fingerprints.
• Bug fixes
• fixed bug in "Docking/De-novo Tools/Scan Fragment" (Windows only)
• fixed bug in aromaticity detection in certain ring systems
• increased number of atom selections ("too many selections created already" error should be more rare)
• fixed bug in Min( alig, seq )
• fixed 2D to 3D conversion macro to avoid getting wrong chiral conformation
• fixed incorrect zoom in the spreadsheet scaffold view.
• Fixed freezing of 3D on Intel graphic cards. (GL_FEEDBACK mode)
• Excel import (importExcel macro) improvements and fixes.
• fixed hydrogen position in build hydrogen command for certain cases

Version 3.8-6a

• Fixes and improvements in pKa models
• Fixed freezing of 3D on Intel graphic cards. (GL_FEEDBACK mode)
• fixed problem in read "https://...."
• Insert/Calculate Column Dialog: 2DFrom_Smiles now correctly works with string column displayed as 2D.
• Fixed freezing in range color dialog on Mac
• Fixed bug in atom charging in Ligand Editor

Version 3.8-6

• added inverse option to set chiral chemical. Inverts chiral centers.
• added sequence accessibility information to alignment view
• Selection propagation between 3D and 2D ligand diagram.
• Table tools panel widgets can be arranged in two rows.
• Advanced column coloring dialog
• Added menu Chemistry.Superposition.Rigid Substructure Superposition to Template.
• fixed issue with fork command. (child may hang in certain cases on exit)
• Support for special values ( > ND, INF etc) in real arrays for Shuffle and show column added.
• fixed bug in set charge formal auto so it does not clear positive charges which come from valency rules.
• Chemical structures can be drawn as labels on plot axis. (smiles or chemical column can be provided)
• Bioisostere search tool supports linker search and replacement
• importExcel macro to import 'xlsx' file with embedded images.
• 'minimize' option to find maximum common substructure command (used in MMP)
• Chemical structures on plot axis labels.
• Rich text support in axis and plot titles (Greek letters, bold, italic, superscript, etc.)
• Pie chart style for both histogram and plot modes.
• New chemical functions for Chemical Properties dialog:
  • pKa of Most basic or acidic group
  • MolCharge negative,positive or total at given pH
• Chemical search dialog work with reaction tables.
• Markush enumeration and filtering: auto completion for filter functions, ability to extract built-in R-groups
• Options to keep header and grid in spreadsheet printing.
• Move table column popup menu
• Metabolic oxidation prediction model (Chemistry/MolScreen)
• Improvements in SALI analysis (bug fixes and plot)
• Improvements in chemical pair analysis (score and plot)
• Find isostere by single atom replacement scan.
• Find bioisostere groups and linkers (right-click chemistry)
• reorder numerical columns by clustering
• basic CDX format import support
• added out-of-bag error estimate for Random Forest prediction model
• improvements and bug fixes in 2D smiles depictions make flat
• align by scaffold improved: make flat added pattern=X_scaffold argument
• template (substucture and apf) support in the Ligand Editor
• added charge option for build tautomer command to enumerate charge states (see also enumerate charge)
• point label size control in plots
• multiple curves in plots
• matched pair analysis (chemMatchedPairs)
• covalent docking hitlist: covalent residue is shown
• Replace( , ) # efficient replace by dictionary
• biomt support for MMTF (read pdb mmtf "4tvp")
• option to show rainbow in chemical cells for color chemical command
• background job finish dialog is non-modal now. (fixes the problem when multiple jobs finish in the same time)

Version 3.8-5a

• fixed bug in read table in CSV or TSV format when large integer numbers were truncated.
• fixed crash in right click in table tools area
• fixed crash when reading some PDB with small default buffer size
• fixed PubChem services (http -> https)
• fixed bug in Structure to Name conversion.

Version 3.8-5

• Collection( ) collection with translation dictionary
• Replace( [] ) translation operation
• delete residue automatically adjusts internal alignment with construct sequence
• Auto close mode for clustering tree (Right click: Preferences)
• Fixes in assign residue (works with cyclic peptides and unusual amino acids)
• residue field "_WORMRADIUS_" for variable thickness ribbon cylinder.
• auto save dir is moved into $s_userDir/bak
• fixed freezing in few places with modal file open dialog on Mac
• selftethers: individual self tethers to a box defined by TOOLS.tsShapeData and Z-only mode
• residue contributions from bonded MMFF terms
• side display/undisplay individual buttons.
• fixed bug in setting sequence type when load PFAM alignment
• new maximum common substructure algorithm. ('sstructure' option for make tree to calculate common substructure for cluster nodes)
• heatmap in plot area
• Initial support for new pdb mmtf format (read pdb mmtf)
• 2D ligand interactions diagrams
• Entropy( R_frequences [ | R_energies r_RT ] ) is added.
• Entropy( seq_ [simple|R_26res_prop] ) is added.
• Added argument to InChi function. (InChI_API_Reference)
• fixed bug in Map() function which may produces broken maps from crystallographic sources.
• rotate and transform translate support linked meshes.
• clustering make tree of very large data sets is possible. (fixed bug in make tree command)
• Score(X_chem_n X_chem_m set ) and Distance( .. ) to calculated similarities and distances between chemical sets
• read gui command to load user-defined menus
• added HTTPS protocol support for read commands
• format for label in the compressed chemical view TOOLS.chemicalGroupSubstFormat
• Added "Edit/Auto Assign Charges" function in the molecular editor
• Added "Modification History" table in the ligand editor
• Control of Z-shift of residue labels (resLabelShift)
• improved coloring by residue type
• alignment panel: ability to select only visible blocks (and not select the hidden ones) with one swipe
• alignment panel: ability to generate pseudo sequences from multiple alignment blocks (e.g. pocket)
• Hetero and R-group scan in the Ligand Editor
• binary read/write for profile objects
• dynamic load of table cell from URL (image or HTML. 'set property load t.col')
• few fixes in chemical depiction
• retina support on Mac (no package yet)
• svg export support for alignment and plots
• read sequence swiss records the 1st res. position after signal peptide or 1 in i_2out
• Cleaned drag'n'drop behavior. ( grob can be dropped on/from objects to link/unlink them )
• 64 bit package for Windows
• Distance( ~r_dist_0to2|M evolution ) returns Dayhoff corrected sequences
• Color( M_Nx3_rgb [ name ] ) # changed to generate hex colors (e.g #ffaaff ), the old mode under the name function. also to support four hex formats for color ( #rgb, #argb, #rrggbb, and #aarrggbb ). formats #argb and #aarrggbb start from the alpha-channel (transparency) value.
• Sequence( ali ) to return a parray of sequences from the alignment; Sequence ( ali group ) - an average sequence. Warning: the default changed. Previously Sequence ( ali ) returned an average sequence.
• Sequence( S_seqNames name ) to return a parray of loaded to ICM sequences by name
• Sequence( S_seqStrings S_seqNames ) create new seq. parray
• Table( s_dot_svg plot ) # parsing an svg file with a graph of nodes and edges into table for resorting
• makeSeqGraph : making 2D sequence maps. macro to generate input for network viewing.
• realignBlock : improvements (allow empty sequences, improve parameters)
• rotateView macro: generating standard view arrays

Version 3.8-4b

• Non R/S chirality is preserved in reactions

Version 3.8-4a

• Fixed memory leak in _dockScan (if ran with -s option)
• Fixed crash in delete sequence from alignment
• Fixed bug in binary write of objects with selftether
• Corrected formal charge calculation for atoms with attachment points
• Fixed bug in _protDesign script

Known bugs in 4a:

• impossible to drop a sequence to an alignment with displayed tree (causes an infinite loop).

Version 3.8-4

• Ligand Editor:
  • Close session function is added
  • Advanced/Export complex improvements
  • New distance and positional restraints
  • Added explicit flexible side-chain support in the Ligand Editor
  • Added an option to minimize ligand in full atom model inside the receptor. (in addition to soft grid minimization)
• Mutation Design:
  Mutation.Protein Protein Binding Mutation.Protein Peptide Binding Mutation.Protein Stability Mutation.Protein Ligand Binding
• GUI Misc
  • Added object popup "Tools.Move.Top/Bottom/Custom" to move objects in the workspace.
  • GUI.workspaceChainGroupStyle and GUI.workspaceChainGroupLimit are introduced to control the chain grouping in the workspace.
  • Feature view and annotation added to single sequences
• New commands/methods
  • Collection( T_tab all ) and Table( k_collection ) functions are added to convert between collections and tables
  • Random Forest classifier and regression are added ( learn chemical )
  • color-molecule-chain(_MAINCOLOR_)
  • undo-store-slidecommand is introduced (push the current display state into undo stack)
  • enumerate-charge command is added to generate various protonation states
  • set-color-alignmentis added
• Misc fixes and improvements:
  • "Docking/Export As Model" (export docking project as a single model icb file)
  • "Docking/Dock Chemical Table" supports models as a project input
  • _dockScan script supports model as project input
  • Several improvements in ECFP fingerprints for prediction models.
  • Fixed background job termination on Windows
  • added bar={ } to the montecarlo command to show the partial progress.
  • Easy molecular placement to a location with optimal distance-specified tethers added
  • Unique color added to molecule
  • Ability to split a pdb object by 'chain' added
  • original object type (XRAY,NMR,etc...) is preserved during the conversion and used in write pdb command
  • fixed bug in linker search when input file is SDF.
  • fixed bug in popup menu from sequence selection in the workspace
  • fixed bug in similarity search in MOLT files.
  • atoms radii are added to slides.
  • compress ali # remove gapped columns
  • Align ( ali dash ) # anneal frayed alignment ends
  • AlignSS ( seq1 seq2 0.1 ) # sequence structure alignment for homology modeling (eg GPCRs)
  • makeAlignmentFromSuper # structural alignment with superposition

Version 3.8-3

• fixed crash in clustering certain large table under Windows (default stack size)
• ND values are ignored in grouping by column (sum,mean,rmsd functions)
• fixed crash in 'delete table' after sorting with distance matrix
• Sort Column Dialog is added.
• Multiple atom coloring schemes in icm.clr file (#COLOR_SCHEME)
• Table copy/paste is improved (name preserved, allows to copy/paste row/column selection)
• Fixed certain cases up/down bonds in 2D->3D conversion.
• Transpose( [ ] ). (Tools/Table/Pivot Column)
• "Test Covalent Mechanisms" is added into 'Chemistry' section of Insert Column Dialog
• add align number chemical command
• non-modal behavior for 2D editor in the 3D Ligand Editor
• Linked meshes moved to the object folder in the workspace
• Improvements in "grouped by column" view: sorting, coloring
• Fixed crash when deleting singleton node in the cluster view
• make-bond-chemicalis added
• Index( connect ) returns iarray with atom numbers marked as attachment points.
• 'Undo' support in protein sculpting mode.

Version 3.8-2

• Fixed crash in certain combination of startup code and file contents
• fixed sorting issues with table is dynamically grouped by a column.
• Added HTTP.headers collection to send custom headers in http request
• Added TCP listener mode to allow remote communications with ICM session.
• Added GRAPHICS.atomRainbow (for cpk and xstick coloring)
• Fixed some issues in http requests with authorization
• Added "Flip Around Bond" operation into Molecule editor
• Fold/unfold columns into array of collection and back
• sort table by field from collection in the array
• editable fields in collection in the table cell

Version 3.8-1

• interactive collection view in tales cells.
• configurable folders in molecular objects: "*so4*;sulfate-ion"
• Score( [similarity|identity|.. [] ] ) extension to permit sequence identity normalized differently. The argument of 0 will divide the score/identity by the alignment length, rather than minimal sequence length. You may also specify a different number of positions.
• calcSeqSimilarity macro for seq-in-alignment relative identities and similarities.
• multiconf MOLT support in _chemSuper
• Added "Stack->Movie" dialog
• Support of multi line annotations in the alignment.
• Added checking the availability of the newer ICM version
• Improved "Edit/Restore Recent Backup" dialog.
• Embedded custom models support.
• Sequence drag'n'drop between sessions.
• Fixed copy/paste reaction issues
• Control of electrostatic calculation with TOOLS.rebelPatchSize (def. 1.) for find-grain calculation
• Smooth electrostatic coloring with the heavy option in color grob potential command.
• a table of 250 non-standard Amino Acids added and enabled for the mutation tool (make it visible with the Tools.AminoAcids menu)
• Chemical dictionary extended for immediate.
• Optimized dynamic group by column for large tables
• Find chemical supports nProc
• compress option for formatted output of real columns in tables
• enumeration with ring formation (macrocycles) in make reaction
• fixed bug in coloring plot by sarray columns
• Added 'hash' option for text columns in plot
• self containing model for APF QSAR
• Added search in PubChem from the general search toolbar
• sort column table is added
• fixed repaint issue of the frozen column in the table.
• _molScreen script and corresponding KNIME node is added. (interface to 2D fingerprint and 3D pharmacophore models)
• SAR analysis command is added.
• R-group decomposition supports R-groups with more than one heavy neighbor. (e.g: R1 in a ring)
• Added MolMechanics/Try Mutation menu for protein binding and protein stability prediction.
• Peptide builder from extended list of amino acids
• Fixed crash on Linux when locale set to some non-English (e.g: de_LU.utf8)

Version 3.8-0

• Export of chemical structures, alignments and plots to Windows Meta-file (WMF) now has anti-aliased fonts and smooth line connections.
• Fixed crash on compress grob command
• Anaglyph stereo mode is added.
• added interface to BLAST search at NCBI (main Search bar in GUI)
• added Extended Connectivity Fingerprints (ECFP)
• Fixed bug with table selection in macros when table with the same name exists globally.
• xstick transparency mask, atom and residue font size and family now are stored in slides.
• Ctrl +/- in the table view resize the cell size (in addition to font size)
• Chemical Search dialog allows to select an existing chemical table for Molcart or local search.
• base html path is set correctly for commands like: read html "http://google.com"
• Chemistry/Full Model Panel is added
• Added macro and GUI to export object's stack as a movie. (Right click on the object: Tools/Export Stack")
• Covalent docking in "batch" mode in the ligand editor.
• "Find in alignment" supports wildcard patterns: N?[TS]
• Added confidence interface visualisation in table plots. (see make plot command and GUI dialog)
• InChI support. InChi( X_chemical [key] ), Chemical( S_inchi )
• Name( user ) returns OS user name.
• "copy off" option for set field command to prevent field from being copied with copy object command.
• Added macro and GUI to export slides as a movie
• added ToxScore function into Insert Column Dialog
• fixed several issues with sugar molecules representation in the workspace
• recursive SMARTS of more than one level are supported.
• row heights are preserved in ICB file
• added GUI interface to add image column and modify its contents
• Fixed crash when deleting 2D label which was used in slides.
• set type mmff on the converted object does not change aromatic bond type to pure aromatic
• added GUI.font
• added support for recursive SMARTS in R-group decomposition ( e.g: [R1;$(CCN)] )
• Distance chemical supports nProc
• Fixed a crash in 2D label edit dialog (image list)
• Image column is exported to Excel.
• Added the ability to add custom functions into 'Insert Column' dialog. (see gui programming for details)
• Dynamic grouping by column
• _chemSuper preserves original columns
• Combined "Pretty" view from docking hit list and Ligand Editor.
• set site alignment and delete site alignment are added
• added simple=S_cols option for read table mol to prevent type guessing (keep column as sarray)
• fixed some issues with HTTP.proxy
• write image alignment and set property alignment commands
• added support for STL mesh format (popular for 3D printing). Example: write grob g_mesh "mesh.stl"
• PubChem CID lookup function and Insert column of pubchem CIDs into a chemical table

Version 3.7-3b

• added search in POCKETOME option to the main search bar
• added clusterTableApf macro
• Wire color is not saved in ICB file. FIXED
• Wire thickness is not saved in ICB file. FIXED
• Fixed inability to drag distance labels after changing font.
• added R/S labels display in the 3D ligand editor
• added optional normal vector argument to Grob( "ELLIPSOID"|"TORUS"|"CYLINDER" ...)
• plot labels now support '\n'
• fixed bug in title label for histogram plot with multiple series.
• icm.gui 'REQUIRED' checks for empty selection
• KNIME nodes v1.01 (Ligand selection in MolDockPrep module, new options in the MolDock module )
• Added Kernel Regression model. (`learn{ lean type="nn" })
• added support for MOL V3000 extension format.
• optimized core replacement search speed
• added blob support in the shell
• UTF8 encoding support in the terminal
• Column format dialog works with column selection.
• OpenMP support nProc variable to toggle internal parallelization of ICM commands.
• Fixed bug in read grob from URL
• Fixed bug in write matrix into ICB file (if rows were reordered)
• Collection( table ) and Table( collection ) were added
• Added the ability to use ICM built-in functions in the function section of the add column function command
• makeIndexPdb macro also builds PDB ligand table which can be search by substructure from the PDB search bar (just paste smiles or use molecular editor)
• Added COLOR preferences (accessible through GUI Preferences dialog)
• Detachable tabs with tables, alignments and other objects. (Drag the tab outside and it will become a separate window).
• Index( alig as_ )
• alig[ I_index ] non-continuous sub-alignment
• added support for optional ph4 features and constraints
• added 'onSelectionChange' object named field to assign actions for selection changes in 3D
• added ability to link mesh to the object for simultaneous rotation, superimpose and display.
• added scanFilterHitlist macro to quickly filter large docking hit lists for top scored hits. (Integrated into main scanMakeHitList macro and corresponding GUI dialog)
• optimized read object for big files.
• added $ support in index expression. (refer to the last element of the array)
• Fixed bug in Date to String conversion. (week of the day and related fields are OK now)
• Added conditional expression support a?b:c
• Added 'delete query' option to modify chemical command to clear SMARTS search attributes.
• logical expression shortcuts: if ( Nof(A)>=5 & A[5] == 1 ) ...
• added 'term' option in macro declaration: saves and restores energy terms after return from the macro
• balloon tooltip for cluster nodes (Right Click/Change Record Labels to specify list of columns in the dialog)
• better docking convergence in the Ligand Editor
• LIGAND.displayDockingMoves is set to 'no' by default
• new Area/Volume distribution for icmPocketFinder plot
• comments for stack conformations. ( set comment store conf Name conf )
• Initial stack of ligand positions before docking in the ligand editor
• 'MolArea' model is added to calculate accessible surface are for 2D chemical. Predict( Chemical("CCCCO") "MolArea" )
• File.ICM Session.New/Clone added.
• JavaScript errors are printed to the terminal
• added Image( X_Chemical )
• covalent ligand support in the ligand editor.
• added triple-quote string constants (arbitrary length and content text can be easily used as string constants)
• make tree object fills "ORDER" atom field which corresponds to the original atom order. (can be used to assign various properties)
• added Chemical( R_react ) and Chemical( X_chem, reaction ) to convert reaction to chemical and back
• write image chemical supports HTML5 canvas export and does not require X11 connection on Linux anymore (can be used in various batch scripts, CGIs, etc. )
• added Info model function to return prediction model properties.
• added Split( [i_minSize] [i_maxSize] bond ) to split molecule into a various fragments
• speed improvements in enumerate tautomer command
• Added String( macro all ) to return macro source and all dependent macros.
• Added String( X_chem html ) String( w_img html ) to export chemicals and images into in-line html representation
• HTML5 canvas support for 2D drawing. (chemicals,plots,alignments,etc. ...)
• read json now accepts entries with multiple roots. (array on the top level)
• XML parser treat comments and CDATA correctly.
• added Select by alignment function
• fixed bug in select alignment
• allow to use ICM user-defined functions in add column function
• _dockScan script is KNIME compatible. Added _dockProjPrep script to prepare docking project in the KNIME work-flow.
• new sugar view in the workspace.
• assign residue command
• Easy selection finder in the alignment ('Highlight Selection' checkbox in the tools panel)
• Balloon popup over plot elements. ('tooltip' option in the command line + GUI element in the dialog). See make plot
• LIGAND.displayDockingMoves option to control display of montecarlo moves in Ligand Editor docking.

Version 3.7-2f Nov 24 2024

• "locked" objects can be copied
• removed '*' from object/mol popup menus

Version 3.7-2e

• fixed rendering of down bond for scaled down compounds
• fixed false interpretation of '$' in single quoted strings
• quotes are processed properly in dialog input elements
• JSON parser accepts array as a top element: Collection( "['a','b']" )
• fixed atom naming in llp residue according to PDB convention
• Interrupt button cancels read from url
• grob clipping works with wire and dot representation.
• Fixed read pdb with 'non-standard' hydrogen names
• Optimized speed of the write map command
• Fixed inverted stereo in problem after "Extract ligand" in some cases.
• Mouse wheel in the workspace scrolls again.
• Fixed display of the variables in the Ligand Editor: Display Strain
• Fixed bug in copy/paste of single table column with empty rows.
• HTML tags from column format are not exported into SDF file

Version 3.7-2d

• Added docking hitlist filtering macro to get N top scored poses. scanFilterHitlist. (The option is also available in the GUI "Make hit list" dialog)
• fixed bug in Formula for structures with no hydrogen.
• Removed unnecessary ligand surface rebuilds in the Ligand Editor
• Fixed bug in display slide when skin representation wasn't preserved.
• Index( X, X atom map function is added to find atom number correspondence between two chemicals
• fixed nested index access to collection
• Fixed crash in make map potential
• Unix and Mac: "Save project As" works if path contains spaces.
• Windows only: -R option sets stdin/stdout to binary mode which makes possible to use pipe-able scripts in KNIME nodes
• fixed numerical instability in minimize
• fixed bug in make reaction
• _confGen and _chemSuper can be used in unix pipes and KNIME nodes
• Min/`Max functions work with date array
• Fixed bug recently introduced in IcmSequence function

Version 3.7-2c

Major new Features:

• New linker and core replacement tool in the 3D ligand editor
• Easy way to write pipe-able scripts (see $ICMHOME/molpipe/*.icm).
• Easy way to add parallelism to unix/mac ICM scripts: fork with pipe option ($ICMHOME\molpipe\*.icm)
• copy/paste and drag'n'drop between icm sessions ( images, molecular objects, meshes, tables )

New Features:

• Added Collection( t|t.column format ) to get the various column properties as a collection object.
• Added Table( model chem [inverse] ) function.
• reading large file by chunks ( read file by chunk ) now supports zipped file.
• Score( X_3Dn [ X_3Dm ] [similarity|distance] ) → M_nxm apf_scores normalization of apf cross-scores.
• added option=s_filename argument for make plot command which allows one to generate plot images in the batch mode.
• added PDF/EPS format for saving plot as an image.
• rename of the object/mesh/map keeps slides and distances in consistent state.
• switching the receptor conformation in the ligand editor rebuild ligand and pocket surface if they are present
• Nof( fork ) is added to get the number of available processors
• pipe options is added for fork command
• a_SLIDE. selection is added. Returns list of objects used in slides, also delete all compress
• learn remembers columns calculated with add column function to automatically recalculated in 'predict and Predict
• _chemSuper supports input from pipe.
• Slides now remember the conformation from the object's stack
• Added stack option for read mol command.
• read mol can read directly from chemical array
• Score( as_ as_ field ) and Score( chem_array field ) are added for APF clustering
• Added stack option for read mol command to read SD file with multiple conformations as an object with a stack.
• Added export alignment to PDF or EPS. (Right click "Save/Export To Image/Save Image")
• Added support of the APF ligand based project in the 3D ligand editor.
• Srmsd( .. .. [weight|chemical] matrix ) returns sarray of static RMSDs or, with option weight, superposition errors computed according to the TOOLS.superimposeMaxDeviation (consistent with r3_out of superimpose minimize )
• Volume( R6 box )
• Table(residue) now returns residues from the icm.res and usr.res libraries. New residues added for CME, CSD, CSO, HYP, KCX, LLP, PTR, SEP, TPO, TYS, and CIR modified amino acids.
• read binary pdb s_4letter_code to read .icb files in pdb-style subdirectories according to the s_xpdbDir root (also extended to allow http , e.g. s_xpdbDir = "http://xablab.ucsd.edu/xpdb/"
• term ts , TOOLS.tsShape and TOOLS.tsShapeData to allow soft penalties for atoms moving outside certain shapes (sphere, box).
• String( s [16|32] key ) (hash function generating 8-character long keys) fixed; String( s hash ) added to generate stronger keys of length 26 or 32
• added coloring for cluster nodes (heat maps)
• added Index compare function to compare different arrays
• parray supported as collection elements
• plot rendering speed is optimized significantly.
• added 'add exact' combination for make grob map command to allow absolute increment
• delete command works directly with index expression in collection (delete c["aaa","bbb"])
• individual atom ball&stick radius.
• added ability to read large portions of data by chunks ( see: read file by chunk )
• added ND word. (for non-defined )
• Added HTTP.cookies collection to allow easier control over http cookies.
• GUI access to molecular named fields (see set field)
• added image rendering for 2D labels

Bug fixes:

• fixed bond removal in cyclopeptides in convert command.
• fixed crash in build tautomer command
• Fixed bug in map contouring with large buffer size values.
• Fixed bug in skin rendering in display stack
• Fixed bug in readUniprotWeb macro when l_references is 'yes'
• Fixed bug in "Filter Graphical Selection" dialog
• Fixed bug in S_ ~ s_ and S_ == s_
• Fixed rare crash in mc
• fixed bug in cis/trans smiles generation ( when cis or trans bond is a ring closure bond )

Version 3.7-2b

• Fixed permission issue in Molcart on MySQL 5.5
• added output= argument for all show commands which allows one to grab the output for further processing.
• Improved rendering speed of xstick representation
• Fixed cis/trans depiction for R-group substituents in enumerate library
• Fixed memory few memory leaks when working with tables in macros.
• Fixed chemical formula generation for isotopes.
• copy/paste multiple cells from Excel into ICM spreadsheet.
• Fixed crash "Changing the final row in a filtered column crashes ICM" bugzilla: #697
• Fixed crash in pasting a column in the table with plot.
• XML parsing functionality. read xml, Collection, xml drugbank example
• fixed incorrect cis/trans bond interpretation in smiles. ( (\R)C=C → R/C=C )
• new attachment point rendering style
• fixed parsing bug in cell hyper links ( set format )
• fixed "interrupted system call" error in read http.
• chiral centers with an R-group attached are correctly treated (stereo is preserved by parity) in make reaction.
• smiles property for sarray columns in a table. Toggles on-the-fly 2D chemical depiction.
• Fixed bug with non-preserving center of rotation in slides after clipping planes were moved. (also fixed in activeICM 1.1-5)
• Multi-receptor support in the 3D editor
• make reaction allows arbitrary R-group numbering
• multiple drag-able resize-able images in set background image command
• added 'Unlock ICB' function which clears read-only or password protection from ICB without reopening it.
• fixed few bugs in with password protected ICB files
• removing salts with modify chem delete salt
• preview for ICB files (Windows browser + ActiveICM)

Version 3.7-2a

• single quote string constant support.
• reader.icm script
• Logical(r|i|s)
• GRAPHICS.cpkClipCaps preference (1,2,3)
• SolveCubic( a,b,c,d | R_3|4 [all] ) → R, SolveQuadratic( a,b,c | R2|3 )
• fixed overlapping shaded and bordered boxes in alignment view. All view preferences now stored in ICB.
• WebKit integration is added
• other types of occlusion shading added ( TOOLS.occlusionColorStyle = "dark outside" or "light outside")
• fixed bug with opposite rotation of connected molecules/objects or grobs
• delete sequence nucleotide|protein|peptide
• fixed rdpdb for consecutive atoms with identical coordinates, e.g. 1a69, ds skin molecule,
• Nof( bond [error]) Nof( selftether [error] ), Select( selftether [error] ), Select( bond [error] )
• selftether treatment: set selftether , delete selftether , Srmsd( as selftether ), minimize "ts" , replaced tether= by selftether= in minimize montecarlo ; convert to set selftether
• improved convertObject (protonation of a_/U, minimize "ts" for the convert deviants)
• added output option for Rmsd and Srmsd function to store individual deviations into R_2out
• grouping by table column works with ND values correctly in aggregation operations.
• write gamess to write correct Z-matrix, memory limits.
• TOOLS.membrane, update icm.hdt, surfaceMethod="membrane" supported in Area( energy) and energy evaluations.
• bug with residue names for residue names starting from a leading blank after reading some pdbs fixed. (previously was breaking 1ytw,2i42,3f9a,2i4e,1kpe)
• CubicRoot(|)
• bug with crashes in find database fixed
• improving find database results (prioritize SC for HUMAN, ECOLI, BOVIN etc. in case of identical scores)
• a_J selection to filter short peptides.
• Name(sequences) returns an empty array (not error) when there are no sequences
• set column format supports internal ICM links.
• exclude and number options for learn atom command.
• set atom named fields is added
• anti-aliased and scalable fonts (atom/residue labels,etc) in 3D for windows and MAC
• Pattern( rs ), selections a_/Bbarcode_like_A12L2L and a_/Qbarcode
• translation to a destination point for a molecular and object selections (see translate )
• new 2D compound rendering option: color rectangles on hetero atoms instead of atom labels. See set property chemical view
• fixed crash in Askg
• new Iarray( a_// topology ) function
• new align sequence command (align number .. )
• new Select_by_sequence function
• find molecule sstructure drestraint ms1_inIcmObj ms2_inTheSameIcmObj added to set drestraints by chemical similarity
• delete link alignment ; link ms_ rewritten, many options added. autoLink action improved.
• join by structure column ignores hydrogens (3D can now be joined with 2D structures). stereo off option is added to ignore chirality.
• copy/paste and drag'n'drop between icm sessions ( images, molecular objects, meshes)
• build tautomer and set tautomer can be applied to HIS residues
• ND support for individual real values
• large 64 bit integer support for integer variables
• Grob( "cylinder", r_ra r_H )
• arbitrary length atom labels and ball radius (`set-label-atom and set atom ball command)
• search functionality in preference dialog.
• restore preference command is added (allows one to set system preferences and variables to default values)
• date type is introduced
• new annotation style for alignments
• Charge/protonation state prediction using pKa model
• pKa predictions for bases and acids (`set-charge-auto command )
• GUI parser patched and icm.gui cleaned from in-line argument declarations to allow natural command syntax (the percent symbol, like %i_out , is no long needed ) in dialog descriptions
• option to avoid sampling His, Gln, Asn, etc. and hydrogens upon convertObject added (allows one to keep atoms as is)
• concurrent multiple level contouring (with multiple colors) of distant density added
• read pdb improved to recognize unusual amino acids
• mmff treatment of two nitrogens in a fused aromatic ring improved
• set field can be used with alignments
• set color alignment is added
• Header(grob)
• make distance append behavior corrected (tool first order). Option make distance append P_atompairs .. added.
• superimpose P_atompairs os_movable command added
• set tether P_atompairs os_movable command added
• Area( grob error ) returns the percentage of unclosed area (to detect surfaces that are not fully closed)
• Volume( grob ) returns the percentage of unclosed area in r_2out
• copyMol, moveMol and jumping molecules
• not properly recognized chemical templates in 2D chem drawing
• unusual amino acids are not recognized ( 2jge )
• mmff treatment of two nitrogens in a fused aromatic ring problematic
• adding ligand (without receptor) to the table causes icm/C5H5N> processLigandSave Error> [2191] index 1 of array out of range [1:0]